Showing 101–150 of 248 results for author: Moe, S

Multi-modal Self-supervised Pre-training for Regulatory Genome Across Cell Types

Authors: Shentong Mo, Xi Fu, Chenyang Hong, Yizhen Chen, Yuxuan Zheng, Xiangru Tang, Zhiqiang Shen, Eric P Xing, Yanyan Lan

Abstract: In the genome biology research, regulatory genome modeling is an important topic for many regulatory downstream tasks, such as promoter classification, transaction factor binding sites prediction. The core problem is to model how regulatory elements interact with each other and its variability across different cell types. However, current deep learning methods often focus on modeling genome sequen… ▽ More In the genome biology research, regulatory genome modeling is an important topic for many regulatory downstream tasks, such as promoter classification, transaction factor binding sites prediction. The core problem is to model how regulatory elements interact with each other and its variability across different cell types. However, current deep learning methods often focus on modeling genome sequences of a fixed set of cell types and do not account for the interaction between multiple regulatory elements, making them only perform well on the cell types in the training set and lack the generalizability required in biological applications. In this work, we propose a simple yet effective approach for pre-training genome data in a multi-modal and self-supervised manner, which we call GeneBERT. Specifically, we simultaneously take the 1d sequence of genome data and a 2d matrix of (transcription factors x regions) as the input, where three pre-training tasks are proposed to improve the robustness and generalizability of our model. We pre-train our model on the ATAC-seq dataset with 17 million genome sequences. We evaluate our GeneBERT on regulatory downstream tasks across different cell types, including promoter classification, transaction factor binding sites prediction, disease risk estimation, and splicing sites prediction. Extensive experiments demonstrate the effectiveness of multi-modal and self-supervised pre-training for large-scale regulatory genomics data. △ Less

Submitted 3 November, 2021; v1 submitted 11 October, 2021; originally announced October 2021.

Learning by Examples Based on Multi-level Optimization

Authors: Shentong Mo, Pengtao Xie

Abstract: Learning by examples, which learns to solve a new problem by looking into how similar problems are solved, is an effective learning method in human learning. When a student learns a new topic, he/she finds out exemplar topics that are similar to this new topic and studies the exemplar topics to deepen the understanding of the new topic. We aim to investigate whether this powerful learning skill ca… ▽ More Learning by examples, which learns to solve a new problem by looking into how similar problems are solved, is an effective learning method in human learning. When a student learns a new topic, he/she finds out exemplar topics that are similar to this new topic and studies the exemplar topics to deepen the understanding of the new topic. We aim to investigate whether this powerful learning skill can be borrowed from humans to improve machine learning as well. In this work, we propose a novel learning approach called Learning By Examples (LBE). Our approach automatically retrieves a set of training examples that are similar to query examples and predicts labels for query examples by using class labels of the retrieved examples. We propose a three-level optimization framework to formulate LBE which involves three stages of learning: learning a Siamese network to retrieve similar examples; learning a matching network to make predictions on query examples by leveraging class labels of retrieved similar examples; learning the ``ground-truth'' similarities between training examples by minimizing the validation loss. We develop an efficient algorithm to solve the LBE problem and conduct extensive experiments on various benchmarks where the results demonstrate the effectiveness of our method on both supervised and few-shot learning. △ Less

Submitted 22 September, 2021; originally announced September 2021.

Kramers-Weyl fermions in the chiral charge density wave material (TaSe$_4$)$_2$I

Authors: Soyeun Kim, Robert C. McKay, Nina Bielinski, Chengxi Zhao, Meng-Kai Lin, Joseph A. Hlevyack, Xuefei Guo, Sung-Kwan Mo, Peter Abbamonte, Tai-Chang Chiang, André Schleife, Daniel P. Shoemaker, Barry Bradlyn, Fahad Mahmood

Abstract: The quasi-one-dimensional chiral charge density wave (CDW) material (TaSe$_4$)$_2$I has been recently predicted to host Kramers-Weyl (KW) fermions which should exist in the vicinity of high symmetry points in the Brillouin zone in chiral materials with strong spin-orbit coupling. However, direct spectroscopic evidence of KW fermions is limited. Here we use helicity-dependent laser-based angle reso… ▽ More The quasi-one-dimensional chiral charge density wave (CDW) material (TaSe$_4$)$_2$I has been recently predicted to host Kramers-Weyl (KW) fermions which should exist in the vicinity of high symmetry points in the Brillouin zone in chiral materials with strong spin-orbit coupling. However, direct spectroscopic evidence of KW fermions is limited. Here we use helicity-dependent laser-based angle resolved photoemission spectroscopy (ARPES) in conjunction with tight-binding and first-principles calculations to identify KW fermions in (TaSe$_4$)$_2$I. We find that topological and symmetry considerations place distinct constraints on the (pseudo-) spin texture and the observed spectra around a KW node. We further reveal an interplay between the spin texture around the chiral KW node and the onset of CDW order in (TaSe$_4$)$_2$I. Our findings highlight the unique topological nature of (TaSe$_4$)$_2$I and provide a pathway for identifying KW fermions in other chiral materials. △ Less

Submitted 24 August, 2021; originally announced August 2021.

Reducing Annotating Load: Active Learning with Synthetic Images in Surgical Instrument Segmentation

Authors: Haonan Peng, Shan Lin, Daniel King, Yun-Hsuan Su, Randall A. Bly, Kris S. Moe, Blake Hannaford

Abstract: Accurate instrument segmentation in endoscopic vision of robot-assisted surgery is challenging due to reflection on the instruments and frequent contacts with tissue. Deep neural networks (DNN) show competitive performance and are in favor in recent years. However, the hunger of DNN for labeled data poses a huge workload of annotation. Motivated by alleviating this workload, we propose a general e… ▽ More Accurate instrument segmentation in endoscopic vision of robot-assisted surgery is challenging due to reflection on the instruments and frequent contacts with tissue. Deep neural networks (DNN) show competitive performance and are in favor in recent years. However, the hunger of DNN for labeled data poses a huge workload of annotation. Motivated by alleviating this workload, we propose a general embeddable method to decrease the usage of labeled real images, using active generated synthetic images. In each active learning iteration, the most informative unlabeled images are first queried by active learning and then labeled. Next, synthetic images are generated based on these selected images. The instruments and backgrounds are cropped out and randomly combined with each other with blending and fusion near the boundary. The effectiveness of the proposed method is validated on 2 sinus surgery datasets and 1 intraabdominal surgery dataset. The results indicate a considerable improvement in performance, especially when the budget for annotation is small. The effectiveness of different types of synthetic images, blending methods, and external background are also studied. All the code is open-sourced at: https://github.com/HaonanPeng/active_syn_generator. △ Less

Submitted 7 August, 2021; originally announced August 2021.

Object-aware Contrastive Learning for Debiased Scene Representation

Authors: Sangwoo Mo, Hyunwoo Kang, Kihyuk Sohn, Chun-Liang Li, **woo Shin

Abstract: Contrastive self-supervised learning has shown impressive results in learning visual representations from unlabeled images by enforcing invariance against different data augmentations. However, the learned representations are often contextually biased to the spurious scene correlations of different objects or object and background, which may harm their generalization on the downstream tasks. To ta… ▽ More Contrastive self-supervised learning has shown impressive results in learning visual representations from unlabeled images by enforcing invariance against different data augmentations. However, the learned representations are often contextually biased to the spurious scene correlations of different objects or object and background, which may harm their generalization on the downstream tasks. To tackle the issue, we develop a novel object-aware contrastive learning framework that first (a) localizes objects in a self-supervised manner and then (b) debias scene correlations via appropriate data augmentations considering the inferred object locations. For (a), we propose the contrastive class activation map (ContraCAM), which finds the most discriminative regions (e.g., objects) in the image compared to the other images using the contrastively trained models. We further improve the ContraCAM to detect multiple objects and entire shapes via an iterative refinement procedure. For (b), we introduce two data augmentations based on ContraCAM, object-aware random crop and background mixup, which reduce contextual and background biases during contrastive self-supervised learning, respectively. Our experiments demonstrate the effectiveness of our representation learning framework, particularly when trained under multi-object images or evaluated under the background (and distribution) shifted images. △ Less

Submitted 26 October, 2021; v1 submitted 30 July, 2021; originally announced August 2021.

Comments: NeurIPS 2021. First two authors contributed equally

Abstract Reasoning via Logic-guided Generation

Authors: Sihyun Yu, Sangwoo Mo, Sungsoo Ahn, **woo Shin

Abstract: Abstract reasoning, i.e., inferring complicated patterns from given observations, is a central building block of artificial general intelligence. While humans find the answer by either eliminating wrong candidates or first constructing the answer, prior deep neural network (DNN)-based methods focus on the former discriminative approach. This paper aims to design a framework for the latter approach… ▽ More Abstract reasoning, i.e., inferring complicated patterns from given observations, is a central building block of artificial general intelligence. While humans find the answer by either eliminating wrong candidates or first constructing the answer, prior deep neural network (DNN)-based methods focus on the former discriminative approach. This paper aims to design a framework for the latter approach and bridge the gap between artificial and human intelligence. To this end, we propose logic-guided generation (LoGe), a novel generative DNN framework that reduces abstract reasoning as an optimization problem in propositional logic. LoGe is composed of three steps: extract propositional variables from images, reason the answer variables with a logic layer, and reconstruct the answer image from the variables. We demonstrate that LoGe outperforms the black box DNN frameworks for generative abstract reasoning under the RAVEN benchmark, i.e., reconstructing answers based on capturing correct rules of various attributes from observations. △ Less

Submitted 11 August, 2021; v1 submitted 22 July, 2021; originally announced July 2021.

Comments: ICML 2021 Workshop on Self-Supervised Learning for Reasoning and Perception (Spotlight Talk)

Delay-Compensated Distributed PDE Control of Traffic with Connected/Automated Vehicles

Authors: Jie Qi, Shurong Mo, Miroslav Krstic

Abstract: We develop an input delay-compensating design for stabilization of an Aw-Rascle-Zhang (ARZ) traffic model in congested regime which is governed by a $2\times 2$ first-order hyperbolic nonlinear PDE. The traffic flow consists of both adaptive cruise control-equipped (ACC-equipped) and manually-driven vehicles. The control input is the time gap of ACC-equipped and connected vehicles, which is subjec… ▽ More We develop an input delay-compensating design for stabilization of an Aw-Rascle-Zhang (ARZ) traffic model in congested regime which is governed by a $2\times 2$ first-order hyperbolic nonlinear PDE. The traffic flow consists of both adaptive cruise control-equipped (ACC-equipped) and manually-driven vehicles. The control input is the time gap of ACC-equipped and connected vehicles, which is subject to delays resulting from communication lag. For the linearized system, a novel three-branch bakcstep** transformation with explicit kernel functions is introduced to compensate the input delay. The transformation is proved æto be bounded, continuous and invertible, with explicit inverse transformation derived. Based on the transformation, we obtain the explicit predictor-feedback controller. We prove exponential stability of the closed-loop system with the delay compensator in $L_2$ norm. The performance improvement of the closed-loop system under the proposed controller is illustrated in simulation. △ Less

Submitted 2 September, 2022; v1 submitted 19 July, 2021; originally announced July 2021.

Electronic structure of superconducting nickelates probed by resonant photoemission spectroscopy

Authors: Zhuoyu Chen, Motoki Osada, Danfeng Li, Emily M. Been, Su-Di Chen, Makoto Hashimoto, Donghui Lu, Sung-Kwan Mo, Kyuho Lee, Bai Yang Wang, Fanny Rodolakis, Jessica L. McChesney, Chun**g Jia, Brian Moritz, Thomas P. Devereaux, Harold Y. Hwang, Zhi-Xun Shen

Abstract: The discovery of infinite-layer nickelate superconductors has spurred enormous interest. While the Ni$^{1+}$ cations possess nominally the same 3$d^9$ configuration as Cu$^{2+}$ in cuprates, the electronic structure variances remain elusive. Here, we present a soft x-ray photoemission spectroscopy study on parent and doped infinite-layer Pr-nickelate thin films with a doped perovskite reference. B… ▽ More The discovery of infinite-layer nickelate superconductors has spurred enormous interest. While the Ni$^{1+}$ cations possess nominally the same 3$d^9$ configuration as Cu$^{2+}$ in cuprates, the electronic structure variances remain elusive. Here, we present a soft x-ray photoemission spectroscopy study on parent and doped infinite-layer Pr-nickelate thin films with a doped perovskite reference. By identifying the Ni character with resonant photoemission and comparison to density functional theory + U (on-site Coulomb repulsion energy) calculations, we estimate U ~5 eV, smaller than the charge transfer energy $Δ$ ~8 eV, confirming the Mott-Hubbard electronic structure in contrast to charge-transfer cuprates. Near the Fermi level ($E_F$), we observe a signature of occupied rare-earth states in the parent compound, which is consistent with a self-do** picture. Our results demonstrate a correlation between the superconducting transition temperature and the oxygen 2$p$ hybridization near $E_F$ when comparing hole-doped nickelates and cuprates. △ Less

Submitted 16 February, 2022; v1 submitted 7 June, 2021; originally announced June 2021.

Comments: 28 pages, 10 figures

Journal ref: Matter 5, 1-10 (2022)

Large bandgap quantum anomalous hall insulator in a designer ferromagnet-topological insulator-ferromagnet heterostructure

Authors: Qile Li, Chi Xuan Trang, Weikang Wu, **woong Hwang, Nikhil Medhekar, Sung-Kwan Mo, Shengyuan A. Yang, Mark T Edmonds

Abstract: Combining magnetism and nontrivial band topology gives rise to quantum anomalous Hall (QAH) insulators and exotic quantum phases such as the QAH effect where current flows without dissipation along quantized edge states. Inducing magnetic order in topological insulators via proximity to a magnetic material offers a promising pathway towards achieving QAH effect at high temperature for lossless tra… ▽ More Combining magnetism and nontrivial band topology gives rise to quantum anomalous Hall (QAH) insulators and exotic quantum phases such as the QAH effect where current flows without dissipation along quantized edge states. Inducing magnetic order in topological insulators via proximity to a magnetic material offers a promising pathway towards achieving QAH effect at high temperature for lossless transport applications. One promising architecture involves a sandwich structure comprising two single layers of MnBi2Te4 (a 2D ferromagnetic insulator) with ultra-thin Bi2Te3 in the middle, and is predicted to yield a robust QAH insulator phase with a bandgap well above thermal energy at room temperature (25 meV). Here we demonstrate the growth of a 1SL MnBi2Te4 / 4QL Bi2Te3 /1SL MnBi2Te4 heterostructure via molecular beam epitaxy, and probe the electronic structure using angle resolved photoelectron spectroscopy. We observe strong hexagonally warped massive Dirac Fermions and a bandgap of 75 meV. The magnetic origin of the gap is confirmed by the observation of broken time reversal symmetry and the exchange-Rashba effect, in excellent agreement with density functional theory calculations. These findings provide insights into magnetic proximity effects in topological insulators, that will move lossless transport in topological insulators towards higher temperature. △ Less

Submitted 18 May, 2021; originally announced May 2021.

Comments: 24 pages

Accurate Mode-Coupling Characterization of Low-Crosstalk Ring-Core Fibers using Integral Calculation based Swept-Wavelength Interferometry Measurement

Authors: Junwei Zhang, Jiangbo Zhu, Junyi Liu, Shuqi Mo, **gxing Zhang, Zhenrui Lin, Lei Shen, Lei Zhang, Jie Luo, Jie Liu, Siyuan Yu

Abstract: In this paper, to accurately characterize the low inter-mode coupling of the weakly-coupled few mode fibers (FMFs), we propose a modified inter-mode coupling characterization method based on swept-wavelength interferometry measurement, in which an integral calculation approach is used to eliminate significant sources of error that may lead to underestimation of the power coupling coefficient. Usin… ▽ More In this paper, to accurately characterize the low inter-mode coupling of the weakly-coupled few mode fibers (FMFs), we propose a modified inter-mode coupling characterization method based on swept-wavelength interferometry measurement, in which an integral calculation approach is used to eliminate significant sources of error that may lead to underestimation of the power coupling coefficient. Using the proposed characterization method, a low-crosstalk ring-core fiber (RCF) with low mode dependent loss (MDL) and with single span length up to 100 km is experimentally measured to have low power coupling coefficients between high-order orbital angular momentum (OAM) mode groups of below -30 dB/km over C band. The measured low coupling coefficients based on the proposed method are verified by the direct system power measurements, proving the feasibility and reliability of the proposed inter-mode coupling characterization method. △ Less

Submitted 29 July, 2021; v1 submitted 8 May, 2021; originally announced May 2021.

Comments: 8 pages, 8 figures

Journal ref: J. Lightw. Technol. 39(2021) 6479-6486

Shoulder Implant X-Ray Manufacturer Classification: Exploring with Vision Transformer

Authors: Meng Zhou, Shanglin Mo

Abstract: Shoulder replacement surgery, also called total shoulder replacement, is a common and complex surgery in Orthopedics discipline. It involves replacing a dead shoulder joint with an artificial implant. In the market, there are many artificial implant manufacturers and each of them may produce different implants with different structures compares to other providers. The problem arises in the followi… ▽ More Shoulder replacement surgery, also called total shoulder replacement, is a common and complex surgery in Orthopedics discipline. It involves replacing a dead shoulder joint with an artificial implant. In the market, there are many artificial implant manufacturers and each of them may produce different implants with different structures compares to other providers. The problem arises in the following situation: a patient has some problems with the shoulder implant accessory and the manufacturer of that implant maybe unknown to either the patient or the doctor, therefore, correctly identification of the manufacturer is the key prior to the treatment. In this paper, we will demonstrate different methods for classifying the manufacturer of a shoulder implant. We will use Vision Transformer approach to this task for the first time ever △ Less

Submitted 21 April, 2021; v1 submitted 15 April, 2021; originally announced April 2021.

Comments: 11 pages, 12 figures

Inherited Weak Topological Insulator Signatures in Topological Hourglass Semimetal Nb3XTe6 (X = Si, Ge)

Authors: Q. Wan, T. Y. Yang, S. Li, M. Yang, Z. Zhu, C. L. Wu, C. Peng, S. K. Mo, W. Wu, Z. H. Chen, Y. B. Huang, L. L. Lev, V. N. Strocov, J. Hu, Z. Q. Mao, Hao Zheng, J. F. Jia, Y. G. Shi, Shengyuan A. Yang, N. Xu

Abstract: Using spin-resolved and angle-resolved photoemission spectroscopy and first-principles calculations, we have identified bulk band inversion and spin polarized surface state evolved from a weak topological insulator (TI) phase in van der Waals materials Nb3XTe6 (X = Si, Ge). The fingerprints of weak TI homologically emerge with hourglass fermions, as multi nodal chains composed by the same pair of… ▽ More Using spin-resolved and angle-resolved photoemission spectroscopy and first-principles calculations, we have identified bulk band inversion and spin polarized surface state evolved from a weak topological insulator (TI) phase in van der Waals materials Nb3XTe6 (X = Si, Ge). The fingerprints of weak TI homologically emerge with hourglass fermions, as multi nodal chains composed by the same pair of valence and conduction bands gapped by spin orbit coupling. The novel topological state, with a pair of valence and conduction bands encoding both weak TI and hourglass semimetal nature, is essential and guaranteed by nonsymmorphic symmetry. It is distinct from TIs studied previously based on band inversions without symmetry protections. △ Less

Submitted 15 April, 2021; originally announced April 2021.

Comments: 4 figures

Journal ref: Phys. Rev. B 103, 165107 (2021)

Anisotropic quasiparticle coherence in nematic BaFe$_2$As$_2$ studied with strain-dependent ARPES

Authors: H. Pfau, S. D. Chen, M. Hashimoto, N. Gauthier, C. R. Rotundu, J. C. Palmstrom, I. R. Fisher, S. -K. Mo, Z. -X. Shen, D. Lu

Abstract: The hallmark of nematic order in iron-based superconductors is a resistivity anisotropy but it is unclear to which extent quasiparticle dispersions, lifetimes and coherence contribute. While the lifted degeneracy of the Fe $d_{xz}$ and $d_{yz}$ dispersions has been studied extensively, only little is known about the two other factors. Here, we combine in situ strain tuning with ARPES and study the… ▽ More The hallmark of nematic order in iron-based superconductors is a resistivity anisotropy but it is unclear to which extent quasiparticle dispersions, lifetimes and coherence contribute. While the lifted degeneracy of the Fe $d_{xz}$ and $d_{yz}$ dispersions has been studied extensively, only little is known about the two other factors. Here, we combine in situ strain tuning with ARPES and study the nematic response of the spectral weight in BaFe$_2$As$_2$. The symmetry analysis of the ARPES spectra demonstrates that the $d_{xz}$ band gains quasiparticle spectral weight compared to the $d_{yz}$ band for negative antisymmetric strain $Δε_{yy}$ suggesting the same response inside the nematic phase. Our results are compatible with a different coherence of the $d_{xz}$ and $d_{yz}$ orbital within a Hund's metal picture. We also discuss the influence of orbital mixing. △ Less

Submitted 10 April, 2021; originally announced April 2021.

Comments: 6 pages, 4 figures

Journal ref: Phys. Rev. B 103, 165136 (2021)

Observation of multi Dirac fermion cloning induced by moiré potential in graphene-SiC heterostructure

Authors: C. L. Wu, Q. Wan, C. Peng, S. K. Mo, R. Z. Li, K. M. Zhao, Y. P. Guo, C. D. Zhang, N. Xu

Abstract: We reexamine the electronic structure of graphene on SiC substrate by angle-resolved photoemission spectroscopy. We directly observed multiply cloning of Dirac cone, in addition to ones previously attributed to reconstruction. The locations, relative distances and anisotropy of Dirac cone replicas fully agree with the moiré pattern of graphene-SiC heterostructure. Our results provide a straightfor… ▽ More We reexamine the electronic structure of graphene on SiC substrate by angle-resolved photoemission spectroscopy. We directly observed multiply cloning of Dirac cone, in addition to ones previously attributed to reconstruction. The locations, relative distances and anisotropy of Dirac cone replicas fully agree with the moiré pattern of graphene-SiC heterostructure. Our results provide a straightforward example of moiré potential modulation in engineering electronic structure with Dirac fermions. △ Less

Submitted 23 March, 2021; originally announced March 2021.

Comments: 4 figures, 1 table

Journal ref: Phys. Rev. B 104, 235130 (2021)

Flatband-Induced Itinerant Ferromagnetism in RbCo$_2$Se$_2$

Authors: Jianwei Huang, Zhicai Wang, Hongsheng Pang, Han Wu, Huibo Cao, Sung-Kwan Mo, Avinash Rustagi, A. F. Kemper, Meng Wang, Ming Yi, R. J. Birgeneau

Abstract: $A$Co$_2$Se$_2$ ($A$=K,Rb,Cs) is a homologue of the iron-based superconductor, $A$Fe$_2$Se$_2$. From a comprehensive study of RbCo$_2$Se$_2… ▽ More $A$Co$_2$Se$_2$ ($A$=K,Rb,Cs) is a homologue of the iron-based superconductor, $A$Fe$_2$Se$_2$. From a comprehensive study of RbCo$_2$Se$_2$ via measurements of magnetization, transport, neutron diffraction, angle-resolved photoemission spectroscopy, and first-principle calculations, we identify a ferromagnetic order accompanied by an orbital-dependent spin-splitting of the electronic dispersions. Furthermore, we identify the ordered moment to be dominated by a $d_{x^2-y^2}$ flatband near the Fermi level, which exhibits the largest spin splitting across the ferromagnetic transition, suggesting an itinerant origin of the ferromagnetism. In the broader context of the iron-based superconductors, we find this $d_{x^2-y^2}$ flatband to be a common feature in the band structures of both iron-chalcogenides and iron-pnictides, accessible via heavy electron do**. △ Less

Submitted 11 March, 2021; originally announced March 2021.

Comments: 8 pages, 7 figures, accepted by Phys. Rev. B

Journal ref: Phys. Rev. B 103, 165105 (2021)

Reinforcement Learning of the Prediction Horizon in Model Predictive Control

Authors: Eivind Bøhn, Sebastien Gros, Signe Moe, Tor Arne Johansen

Abstract: Model predictive control (MPC) is a powerful trajectory optimization control technique capable of controlling complex nonlinear systems while respecting system constraints and ensuring safe operation. The MPC's capabilities come at the cost of a high online computational complexity, the requirement of an accurate model of the system dynamics, and the necessity of tuning its parameters to the speci… ▽ More Model predictive control (MPC) is a powerful trajectory optimization control technique capable of controlling complex nonlinear systems while respecting system constraints and ensuring safe operation. The MPC's capabilities come at the cost of a high online computational complexity, the requirement of an accurate model of the system dynamics, and the necessity of tuning its parameters to the specific control application. The main tunable parameter affecting the computational complexity is the prediction horizon length, controlling how far into the future the MPC predicts the system response and thus evaluates the optimality of its computed trajectory. A longer horizon generally increases the control performance, but requires an increasingly powerful computing platform, excluding certain control applications.The performance sensitivity to the prediction horizon length varies over the state space, and this motivated the adaptive horizon model predictive control (AHMPC), which adapts the prediction horizon according to some criteria. In this paper we propose to learn the optimal prediction horizon as a function of the state using reinforcement learning (RL). We show how the RL learning problem can be formulated and test our method on two control tasks, showing clear improvements over the fixed horizon MPC scheme, while requiring only minutes of learning. △ Less

Submitted 22 February, 2021; originally announced February 2021.

Comments: This work has been submitted to IFAC NMPC 2021 for possible publication

Improving prediction of the terrestrial water storage anomalies during the GRACE and GRACE-FO gap with Bayesian convolutional neural networks

Authors: Shaoxing Mo, Yulong Zhong, Xiaoqing Shi, Wei Feng, Xin Yin, Jichun Wu

Abstract: The Gravity Recovery and Climate Experiment (GRACE) satellite and its successor GRACE Follow-On (GRACE-FO) provide valuable and accurate observations of terrestrial water storage anomalies (TWSAs) at a global scale. However, there is an approximately one-year observation gap of TWSAs between GRACE and GRACE-FO. This poses a challenge for practical applications, as discontinuity in the TWSA observa… ▽ More The Gravity Recovery and Climate Experiment (GRACE) satellite and its successor GRACE Follow-On (GRACE-FO) provide valuable and accurate observations of terrestrial water storage anomalies (TWSAs) at a global scale. However, there is an approximately one-year observation gap of TWSAs between GRACE and GRACE-FO. This poses a challenge for practical applications, as discontinuity in the TWSA observations may introduce significant biases and uncertainties in the hydrological model predictions and consequently mislead decision making. To tackle this challenge, a Bayesian convolutional neural network (BCNN) driven by climatic data is proposed in this study to bridge this gap at a global scale. Enhanced by integrating recent advances in deep learning, including the attention mechanisms and the residual and dense connections, BCNN can automatically and efficiently extract important features for TWSA predictions from multi-source input data. The predicted TWSAs are compared to the hydrological model outputs and three recent TWSA prediction products. The comparison suggests the superior performance of BCNN in providing improved predictions of TWSAs during the gap in particular in the relatively arid regions. The BCNN's ability to identify the extreme dry and wet events during the gap period is further discussed and comprehensively demonstrated by comparing with the precipitation anomalies, drought index, ground/surface water levels. Results indicate that BCNN is capable of offering a reliable solution to maintain the TWSA data continuity and quantify the impacts of climate extremes during the gap. △ Less

Submitted 7 March, 2021; v1 submitted 21 January, 2021; originally announced January 2021.

Comments: 27 pages, 13 figures

Observation of a Smoothly Tunable Dirac Point in $Ge(Bi_{x}Sb_{1-x})_{2}Te_{4}$

Authors: Sean Howard, Arjun Raghavan, Davide Iaia, Caizhi Xu, David Flötotto, Man-Hong Wong, Sung-Kwan Mo, Bahadur Singh, Raman Sankar, Hsin Lin, Tai-Chang Chiang, Vidya Madhavan

Abstract: State-of-the-art topological devices require the use topological surface states to drive electronic transport. In this study, we examine a tunable topological system, $Ge(Bi_{x}Sb_{1-x})_{2}Te_{4}$, for a range of 'x' values from 0 to 1, using a combination of Fourier Transform Scanning Tunneling Spectroscopy (FT-STS) and Angle-Resolved Photoemission Spectroscopy (ARPES). Our results show that the… ▽ More State-of-the-art topological devices require the use topological surface states to drive electronic transport. In this study, we examine a tunable topological system, $Ge(Bi_{x}Sb_{1-x})_{2}Te_{4}$, for a range of 'x' values from 0 to 1, using a combination of Fourier Transform Scanning Tunneling Spectroscopy (FT-STS) and Angle-Resolved Photoemission Spectroscopy (ARPES). Our results show that the Dirac point shifts linearly with 'x', crossing the Fermi energy near x = 0.7. This novel observation of a smoothly tunable, isolated Dirac point crossing through the topological transport regime and having strong linear dependence with substitution can be critical for future topological spintronics applications. △ Less

Submitted 21 October, 2021; v1 submitted 13 January, 2021; originally announced January 2021.

Comments: 18 Pages, 9 Figures, including Appendix

Journal ref: Phys. Rev. Mater. 6, 044201 (2022)

MASKER: Masked Keyword Regularization for Reliable Text Classification

Authors: Seung Jun Moon, Sangwoo Mo, Kimin Lee, Jaeho Lee, **woo Shin

Abstract: Pre-trained language models have achieved state-of-the-art accuracies on various text classification tasks, e.g., sentiment analysis, natural language inference, and semantic textual similarity. However, the reliability of the fine-tuned text classifiers is an often underlooked performance criterion. For instance, one may desire a model that can detect out-of-distribution (OOD) samples (drawn far… ▽ More Pre-trained language models have achieved state-of-the-art accuracies on various text classification tasks, e.g., sentiment analysis, natural language inference, and semantic textual similarity. However, the reliability of the fine-tuned text classifiers is an often underlooked performance criterion. For instance, one may desire a model that can detect out-of-distribution (OOD) samples (drawn far from training distribution) or be robust against domain shifts. We claim that one central obstacle to the reliability is the over-reliance of the model on a limited number of keywords, instead of looking at the whole context. In particular, we find that (a) OOD samples often contain in-distribution keywords, while (b) cross-domain samples may not always contain keywords; over-relying on the keywords can be problematic for both cases. In light of this observation, we propose a simple yet effective fine-tuning method, coined masked keyword regularization (MASKER), that facilitates context-based prediction. MASKER regularizes the model to reconstruct the keywords from the rest of the words and make low-confidence predictions without enough context. When applied to various pre-trained language models (e.g., BERT, RoBERTa, and ALBERT), we demonstrate that MASKER improves OOD detection and cross-domain generalization without degrading classification accuracy. Code is available at https://github.com/alinlab/MASKER. △ Less

Submitted 16 December, 2020; originally announced December 2020.

Comments: AAAI 2021. First two authors contributed equally

Towards Improving Spatiotemporal Action Recognition in Videos

Authors: Shentong Mo, Xiaoqing Tan, **gfei Xia, Pinxu Ren

Abstract: Spatiotemporal action recognition deals with locating and classifying actions in videos. Motivated by the latest state-of-the-art real-time object detector You Only Watch Once (YOWO), we aim to modify its structure to increase action detection precision and reduce computational time. Specifically, we propose four novel approaches in attempts to improve YOWO and address the imbalanced class issue i… ▽ More Spatiotemporal action recognition deals with locating and classifying actions in videos. Motivated by the latest state-of-the-art real-time object detector You Only Watch Once (YOWO), we aim to modify its structure to increase action detection precision and reduce computational time. Specifically, we propose four novel approaches in attempts to improve YOWO and address the imbalanced class issue in videos by modifying the loss function. We consider two moderate-sized datasets to apply our modification of YOWO - the popular Joint-annotated Human Motion Data Base (J-HMDB-21) and a private dataset of restaurant video footage provided by a Carnegie Mellon University-based startup, Agot.AI. The latter involves fast-moving actions with small objects as well as unbalanced data classes, making the task of action localization more challenging. We implement our proposed methods in the GitHub repository https://github.com/stoneMo/YOWOv2. △ Less

Submitted 15 December, 2020; originally announced December 2020.

Automatic Speech Verification Spoofing Detection

Authors: Shentong Mo, Haofan Wang, Pinxu Ren, Ta-Chung Chi

Abstract: Automatic speech verification (ASV) is the technology to determine the identity of a person based on their voice. While being convenient for identity verification, we should aim for the highest system security standard given that it is the safeguard of valuable digital assets. Bearing this in mind, we follow the setup in ASVSpoof 2019 competition to develop potential countermeasures that are robus… ▽ More Automatic speech verification (ASV) is the technology to determine the identity of a person based on their voice. While being convenient for identity verification, we should aim for the highest system security standard given that it is the safeguard of valuable digital assets. Bearing this in mind, we follow the setup in ASVSpoof 2019 competition to develop potential countermeasures that are robust and efficient. Two metrics, EER and t-DCF, will be used for system evaluation. △ Less

Submitted 15 December, 2020; originally announced December 2020.

Optimization of the Model Predictive Control Update Interval Using Reinforcement Learning

Authors: Eivind Bøhn, Sebastien Gros, Signe Moe, Tor Arne Johansen

Abstract: In control applications there is often a compromise that needs to be made with regards to the complexity and performance of the controller and the computational resources that are available. For instance, the typical hardware platform in embedded control applications is a microcontroller with limited memory and processing power, and for battery powered applications the control system can account f… ▽ More In control applications there is often a compromise that needs to be made with regards to the complexity and performance of the controller and the computational resources that are available. For instance, the typical hardware platform in embedded control applications is a microcontroller with limited memory and processing power, and for battery powered applications the control system can account for a significant portion of the energy consumption. We propose a controller architecture in which the computational cost is explicitly optimized along with the control objective. This is achieved by a three-part architecture where a high-level, computationally expensive controller generates plans, which a computationally simpler controller executes by compensating for prediction errors, while a recomputation policy decides when the plan should be recomputed. In this paper, we employ model predictive control (MPC) as the high-level plan-generating controller, a linear state feedback controller as the simpler compensating controller, and reinforcement learning (RL) to learn the recomputation policy. Simulation results for two examples showcase the architecture's ability to improve upon the MPC approach and find reasonable compromises weighing the performance on the control objective and the computational resources expended. △ Less

Submitted 26 November, 2020; originally announced November 2020.

Comments: Submitted to 3rd Annual Learning for Dynamics and Control Conference (L4DC 2021)

Multi-frame Feature Aggregation for Real-time Instrument Segmentation in Endoscopic Video

Authors: Shan Lin, Fangbo Qin, Haonan Peng, Randall A. Bly, Kris S. Moe, Blake Hannaford

Abstract: Deep learning-based methods have achieved promising results on surgical instrument segmentation. However, the high computation cost may limit the application of deep models to time-sensitive tasks such as online surgical video analysis for robotic-assisted surgery. Moreover, current methods may still suffer from challenging conditions in surgical images such as various lighting conditions and the… ▽ More Deep learning-based methods have achieved promising results on surgical instrument segmentation. However, the high computation cost may limit the application of deep models to time-sensitive tasks such as online surgical video analysis for robotic-assisted surgery. Moreover, current methods may still suffer from challenging conditions in surgical images such as various lighting conditions and the presence of blood. We propose a novel Multi-frame Feature Aggregation (MFFA) module to aggregate video frame features temporally and spatially in a recurrent mode. By distributing the computation load of deep feature extraction over sequential frames, we can use a lightweight encoder to reduce the computation costs at each time step. Moreover, public surgical videos usually are not labeled frame by frame, so we develop a method that can randomly synthesize a surgical frame sequence from a single labeled frame to assist network training. We demonstrate that our approach achieves superior performance to corresponding deeper segmentation models on two public surgery datasets. △ Less

Submitted 25 July, 2021; v1 submitted 17 November, 2020; originally announced November 2020.

Comments: Published in IEEE Robotics and Automation Letters (Early Access)

Non-Thermal Emergence of an Orbital-Selective Mott Phase in FeTe$_{1-x}$Se$_x$

Authors: Jianwei Huang, Rong Yu, Zhijun Xu, Jian-Xin Zhu, Qianni Jiang, Meng Wang, Han Wu, Tong Chen, Jonathan D. Denlinger, Sung-Kwan Mo, Makoto Hashimoto, Genda Gu, Pengcheng Dai, Jiun-Haw Chu, Donghui Lu, Qimiao Si, Robert J. Birgeneau, M. Yi

Abstract: Electronic correlation is of fundamental importance to high temperature superconductivity. Iron-based superconductors are believed to possess moderate correlation strength, which combined with their multi-orbital nature makes them a fascinating platform for the emergence of exotic phenomena. A particularly striking form is the emergence of an orbital selective Mott phase, where the localization of… ▽ More Electronic correlation is of fundamental importance to high temperature superconductivity. Iron-based superconductors are believed to possess moderate correlation strength, which combined with their multi-orbital nature makes them a fascinating platform for the emergence of exotic phenomena. A particularly striking form is the emergence of an orbital selective Mott phase, where the localization of a subset of orbitals leads to a drastically reconstructed Fermi surface. Here, we report spectroscopic evidence of the reorganization of the Fermi surface from FeSe to FeTe as Se is substituted by Te. We uncover a particularly transparent way to visualize the localization of the $d_{xy}$ electron orbital through the suppression of its hybridization with the more coherent $d$ electron orbitals, which leads to a redistribution of the orbital-dependent spectral weight near the Fermi level. These noteworthy features of the Fermi surface are accompanied by a divergent behavior of a band renormalization in the $d_{xy}$ orbital. All of our observations are further supported by our theoretical calculations to be salient spectroscopic signatures of such a non-thermal evolution from a strongly correlated metallic phase towards an orbital-selective Mott phase in FeTe$_{1-x}$Se$_x$ as Se concentration is reduced. △ Less

Submitted 25 February, 2021; v1 submitted 26 October, 2020; originally announced October 2020.

Comments: 11 pages, 5 figures

Journal ref: Commun Phys 5, 29 (2022)

Layer-adaptive sparsity for the Magnitude-based Pruning

Authors: Jaeho Lee, Sejun Park, Sangwoo Mo, Sungsoo Ahn, **woo Shin

Abstract: Recent discoveries on neural network pruning reveal that, with a carefully chosen layerwise sparsity, a simple magnitude-based pruning achieves state-of-the-art tradeoff between sparsity and performance. However, without a clear consensus on "how to choose," the layerwise sparsities are mostly selected algorithm-by-algorithm, often resorting to handcrafted heuristics or an extensive hyperparameter… ▽ More Recent discoveries on neural network pruning reveal that, with a carefully chosen layerwise sparsity, a simple magnitude-based pruning achieves state-of-the-art tradeoff between sparsity and performance. However, without a clear consensus on "how to choose," the layerwise sparsities are mostly selected algorithm-by-algorithm, often resorting to handcrafted heuristics or an extensive hyperparameter search. To fill this gap, we propose a novel importance score for global pruning, coined layer-adaptive magnitude-based pruning (LAMP) score; the score is a rescaled version of weight magnitude that incorporates the model-level $\ell_2$ distortion incurred by pruning, and does not require any hyperparameter tuning or heavy computation. Under various image classification setups, LAMP consistently outperforms popular existing schemes for layerwise sparsity selection. Furthermore, we observe that LAMP continues to outperform baselines even in weight-rewinding setups, while the connectivity-oriented layerwise sparsity (the strongest baseline overall) performs worse than a simple global magnitude-based pruning in this case. Code: https://github.com/jaeho-lee/layer-adaptive-sparsity △ Less

Submitted 9 May, 2021; v1 submitted 15 October, 2020; originally announced October 2020.

Comments: ICLR 2021. Changed title (previous ver: A deeper look at the layerwise sparsity of magnitude-based pruning)

Imaging spinon density modulations in a 2D quantum spin liquid

Authors: Wei Ruan, Yi Chen, Shujie Tang, **woong Hwang, Hsin-Zon Tsai, Ryan Lee, Meng Wu, Hye** Ryu, Salman Kahn, Franklin Liou, Caihong Jia, Andrew Aikawa, Choongyu Hwang, Feng Wang, Yongseong Choi, Steven G. Louie, Patrick A. Lee, Zhi-Xun Shen, Sung-Kwan Mo, Michael F. Crommie

Abstract: Two-dimensional triangular-lattice antiferromagnets are predicted under some conditions to exhibit a quantum spin liquid ground state whose low-energy behavior is described by a spinon Fermi surface. Directly imaging the resulting spinons, however, is difficult due to their fractional, chargeless nature. Here we use scanning tunneling spectroscopy to image spinon density modulations arising from a… ▽ More Two-dimensional triangular-lattice antiferromagnets are predicted under some conditions to exhibit a quantum spin liquid ground state whose low-energy behavior is described by a spinon Fermi surface. Directly imaging the resulting spinons, however, is difficult due to their fractional, chargeless nature. Here we use scanning tunneling spectroscopy to image spinon density modulations arising from a spinon Fermi surface instability in single-layer 1T-TaSe$_2$, a two-dimensional Mott insulator. We first demonstrate the existence of localized spins arranged on a triangular lattice in single-layer 1T-TaSe$_2$ by contacting it to a metallic 1H-TaSe$_2$ layer and measuring the Kondo effect. Subsequent spectroscopic imaging of isolated, single-layer 1T-TaSe$_2$ reveals long-wavelength modulations at Hubbard band energies that reflect spinon density modulations. This allows direct experimental measurement of the spinon Fermi wavevector, in good agreement with theoretical predictions for a 2D quantum spin liquid. These results establish single-layer 1T-TaSe$_2$ as a new platform for studying novel two-dimensional quantum-spin-liquid phenomena. △ Less

Submitted 15 September, 2020; originally announced September 2020.

Crossover from 2D ferromagnetic insulator to wide bandgap quantum anomalous Hall insulator in ultra-thin MnBi2Te4

Authors: Chi Xuan Trang, Qile Li, Yuefeng Yin, **woong Hwang, Golrokh Akhgar, Iolanda Di Bernardo, Antonija Grubišić-Čabo, Anton Tadich, Michael S. Fuhrer, Sung- Kwan Mo, Nikhil Medhekar, Mark T. Edmonds

Abstract: Intrinsic magnetic topological insulators offer low disorder and large magnetic bandgaps for robust magnetic topological phases operating at higher temperatures. By controlling the layer thickness, emergent phenomena such as the Quantum Anomalous Hall (QAH) effect and axion insulator phases have been realised. These observations occur at temperatures significantly lower than the Neel temperature o… ▽ More Intrinsic magnetic topological insulators offer low disorder and large magnetic bandgaps for robust magnetic topological phases operating at higher temperatures. By controlling the layer thickness, emergent phenomena such as the Quantum Anomalous Hall (QAH) effect and axion insulator phases have been realised. These observations occur at temperatures significantly lower than the Neel temperature of bulk MnBi2Te4, and measurement of the magnetic energy gap at the Dirac point in ultra-thin MnBi2Te4 has yet to be achieved. Critical to achieving the promise of this system is a direct measurement of the layer-dependent energy gap and verifying whether the gap is magnetic in the QAH phase. Here we utilise temperature dependent angle-resolved photoemission spectroscopy to study epitaxial ultra-thin MnBi2Te4. We directly observe a layer dependent crossover from a 2D ferromagnetic insulator with a bandgap greater than 780 meV in one septuple layer (1 SL) to a QAH insulator with a large energy gap (>100 meV) at 8 K in 3 and 5 SL MnBi2Te4. The QAH gap is confirmed to be magnetic in origin, as it abruptly diminishes with increasing temperature above 8 K. The direct observation of a large magnetic energy gap in the QAH phase of few-SL MnBi2Te4 is promising for further increasing the operating temperature of QAH materials. △ Less

Submitted 16 March, 2021; v1 submitted 13 September, 2020; originally announced September 2020.

Journal ref: ACS Nano 2021, 15, 8, 13444-13452

Progress in epitaxial thin-film Na3Bi as a topological electronic material

Authors: I. Di Bernardo, J. Hellerstedt, C. Liu, G. Akhgar, W. Wu, S. A. Yang, D. Culcer, S. -K. Mo, S. Adam, M. T. Edmonds, M. S. Fuhrer

Abstract: Na3Bi was the first experimentally verified topological Dirac semimetal (TDS), and is a 3D analogue of graphene hosting relativistic Dirac fermions. Its unconventional momentum-energy relationship is interesting from a fundamental perspective, yielding exciting physical properties such as chiral charge carriers, the chiral anomaly, and weak anti-localization. It also shows promise for realising to… ▽ More Na3Bi was the first experimentally verified topological Dirac semimetal (TDS), and is a 3D analogue of graphene hosting relativistic Dirac fermions. Its unconventional momentum-energy relationship is interesting from a fundamental perspective, yielding exciting physical properties such as chiral charge carriers, the chiral anomaly, and weak anti-localization. It also shows promise for realising topological electronic devices such as topological transistors. In this review, an overview of the substantial progress achieved in the last few years on Na3Bi is presented, with a focus on technologically relevant large-area thin films synthesised via molecular beam epitaxy. Key theoretical aspects underpinning the unique electronic properties of Na3Bi are introduced. Next, the growth process on different substrates is reviewed. Spectroscopic and microscopic features are illustrated, and an analysis of semi-classical and quantum transport phenomena in different do** regimes is provided. The emergent properties arising from confinement in two dimensions, including thickness-dependent and electric-field driven topological phase transitions, are addressed, with an outlook towards current challenges and expected future progress. △ Less

Submitted 1 September, 2020; originally announced September 2020.

Proximity-induced hidden order transition in a correlated heterostructure Sr$_2$VO$_3$FeAs

Authors: Sunghun Kim, Jong Mok Ok, Hanbit Oh, Chang-il Kwon, Y. Zhang, J. D. Denlinger, S. -K. Mo, F. Wolff-Fabris, E. Kampert, Eun-Gook Moon, C. Kim, Jun Sung Kim, Y. K. Kim

Abstract: Symmetry is one of the most significant concepts in physics, and its importance has been largely manifested in phase transitions by its spontaneous breaking. In strongly correlated systems, however, mysterious and enigmatic phase transitions, inapplicable of the symmetry description, have been discovered and often dubbed hidden order transitions, as found in, $\it{e.g.}$, high-$T_C$ cuprates, heav… ▽ More Symmetry is one of the most significant concepts in physics, and its importance has been largely manifested in phase transitions by its spontaneous breaking. In strongly correlated systems, however, mysterious and enigmatic phase transitions, inapplicable of the symmetry description, have been discovered and often dubbed hidden order transitions, as found in, $\it{e.g.}$, high-$T_C$ cuprates, heavy fermion superconductors, and quantum spin liquid candidates. Here, we report a new type of hidden order transition in a correlated heterostructure Sr$_2$VO$_3$FeAs, whose origin is attributed to an unusually enhanced Kondo-type proximity coupling between localized spins of V and itinerant electrons of FeAs. Most notably, a fully isotropic gap opening, identified by angle-resolved photoemission spectroscopy, occurs selectively in one of the Fermi surfaces below $T_{\rm HO}$ $\sim$ 150 K, associated with a singular behavior of the specific heat and a strong enhancement on the anisotropic magnetoresistance. These observations are incompatible with the prevalent broken-symmetry-driven scenarios of electronic gap opening and highlight a critical role of proximity coupling. Our findings demonstrate that correlated heterostructures offer a novel platform for design and engineering of exotic hidden order phases. △ Less

Submitted 27 August, 2020; originally announced August 2020.

Comments: 8 pages with 4 figures

The nature of ferromagnetism in the chiral helimagnet $Cr_{1/3}NbS_{2}$

Authors: N. Sirica, P. Vilmercati, F. Bondino, I. Pis, S. Nappini, S. -K. Mo, A. V. Fedorov, P. K. Das, I. Vobornik, J. Fujii, L. Li, D. Sapkota, D. S. Parker, D. G. Mandrus, N. Mannella

Abstract: The chiral helimagnet, $Cr_{1/3}NbS_{2}$, hosts exotic spin textures, whose influence on the magneto-transport properties, make this material an ideal candidate for future spintronic applications. To date, the interplay between macroscopic magnetic and transport degrees of freedom is believed to result from a reduction in carrier scattering following spin order. Here, we present electronic structu… ▽ More The chiral helimagnet, $Cr_{1/3}NbS_{2}$, hosts exotic spin textures, whose influence on the magneto-transport properties, make this material an ideal candidate for future spintronic applications. To date, the interplay between macroscopic magnetic and transport degrees of freedom is believed to result from a reduction in carrier scattering following spin order. Here, we present electronic structure measurements through the helimagnetic transition temperature, $T_{C}$ that challenges this view by showing a Fermi surface comprised of strongly hybridized Nb- and Cr- derived electronic states, and spectral weight in proximity to the Fermi level to anomalously increases as temperature is lowered below $T_{C}$. These findings are rationalized on the basis of first principle, density functional theory calculations, which reveal a large nearest-neighbor exchange energy, suggesting the interaction between local spin moments and hybridized Nb- and Cr- derived itinerant states to go beyond the perturbative interaction of Ruderman-Kittel-Kasuya-Yosida, suggesting instead a mechanism rooted in a Hund's exchange interaction. △ Less

Submitted 17 August, 2020; originally announced August 2020.

Journal ref: Communications Physics 3, 65 (2020)

A plausible method of preparing the ideal p-n junction interface of a thermoelectric material by surface do**

Authors: Ji-Eun Lee, **woong Hwang, Minhee Kang, Hyun-Jeong Joo, Hye** Ryu, Kyoo Kim, Yongsam Kim, Namdong Kim, Anh Tuan Duong, Sunglae Cho, Sung-Kwan Mo, Choongyu Hwang, Imjeong Ho-Soon Yang

Abstract: Recent advances in two-dimensional (2D) crystals make it possible to realize an ideal interface structure that is required for device applications. Specifically, a p-n junction made of 2D crystals is predicted to exhibit an atomically well-defined interface that will lead to high device performance. Using angle-resolved photoemission spectroscopy, a simple surface treatment was shown to allow the… ▽ More Recent advances in two-dimensional (2D) crystals make it possible to realize an ideal interface structure that is required for device applications. Specifically, a p-n junction made of 2D crystals is predicted to exhibit an atomically well-defined interface that will lead to high device performance. Using angle-resolved photoemission spectroscopy, a simple surface treatment was shown to allow the possible formation of such an interface. Ta adsorption on the surface of a p-doped SnSe shifts the valence band maximum towards higher binding energy due to the charge transfer from Ta to SnSe that is highly localized at the surface due to the layered structure of SnSe. As a result, the charge carriers of the surface are changed from holes of its bulk characteristics to electrons, while the bulk remains as a p-type semiconductor. This observation suggests that the well-defined interface of a p-n junction with an atomically thin {\it n}-region is formed between Ta-adsorbed surface and bulk. △ Less

Submitted 18 July, 2020; originally announced July 2020.

Comments: 4 figures

Journal ref: Appl. Surf. Sci. 520, 146314 (2020)

CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances

Authors: Jihoon Tack, Sangwoo Mo, Jongheon Jeong, **woo Shin

Abstract: Novelty detection, i.e., identifying whether a given sample is drawn from outside the training distribution, is essential for reliable machine learning. To this end, there have been many attempts at learning a representation well-suited for novelty detection and designing a score based on such representation. In this paper, we propose a simple, yet effective method named contrasting shifted instan… ▽ More Novelty detection, i.e., identifying whether a given sample is drawn from outside the training distribution, is essential for reliable machine learning. To this end, there have been many attempts at learning a representation well-suited for novelty detection and designing a score based on such representation. In this paper, we propose a simple, yet effective method named contrasting shifted instances (CSI), inspired by the recent success on contrastive learning of visual representations. Specifically, in addition to contrasting a given sample with other instances as in conventional contrastive learning methods, our training scheme contrasts the sample with distributionally-shifted augmentations of itself. Based on this, we propose a new detection score that is specific to the proposed training scheme. Our experiments demonstrate the superiority of our method under various novelty detection scenarios, including unlabeled one-class, unlabeled multi-class and labeled multi-class settings, with various image benchmark datasets. Code and pre-trained models are available at https://github.com/alinlab/CSI. △ Less

Submitted 21 October, 2020; v1 submitted 16 July, 2020; originally announced July 2020.

Comments: NeurIPS 2020. First two authors contributed equally

Bus Frequency Optimization: When Waiting Time Matters in User Satisfaction

Authors: Songsong Mo, Zhifeng Bao, Baihua Zheng, Zhiyong Peng

Abstract: Reorganizing bus frequency to cater for the actual travel demand can save the cost of the public transport system significantly. Many, if not all, existing studies formulate this as a bus frequency optimization problem which tries to minimize passengers' average waiting time. However, many investigations have confirmed that the user satisfaction drops faster as the waiting time increases. Conseque… ▽ More Reorganizing bus frequency to cater for the actual travel demand can save the cost of the public transport system significantly. Many, if not all, existing studies formulate this as a bus frequency optimization problem which tries to minimize passengers' average waiting time. However, many investigations have confirmed that the user satisfaction drops faster as the waiting time increases. Consequently, this paper studies the bus frequency optimization problem considering the user satisfaction. Specifically, for the first time to our best knowledge, we study how to schedule the buses such that the total number of passengers who could receive their bus services within the waiting time threshold is maximized. We prove that this problem is NP-hard, and present an index-based algorithm with $(1-1/e)$ approximation ratio. By exploiting the locality property of routes in a bus network, we propose a partition-based greedy method which achieves a $(1-ρ)(1-1/e)$ approximation ratio. Then we propose a progressive partition-based greedy method to further improve the efficiency while achieving a $(1-ρ)(1-1/e-\varepsilon)$ approximation ratio. Experiments on a real city-wide bus dataset in Singapore verify the efficiency, effectiveness, and scalability of our methods. △ Less

Submitted 23 March, 2020; originally announced April 2020.

Journal ref: International Conference on Database Systems for Advanced Applications 2020

LC-GAN: Image-to-image Translation Based on Generative Adversarial Network for Endoscopic Images

Authors: Shan Lin, Fangbo Qin, Yangming Li, Randall A. Bly, Kris S. Moe, Blake Hannaford

Abstract: Intelligent vision is appealing in computer-assisted and robotic surgeries. Vision-based analysis with deep learning usually requires large labeled datasets, but manual data labeling is expensive and time-consuming in medical problems. We investigate a novel cross-domain strategy to reduce the need for manual data labeling by proposing an image-to-image translation model live-cadaver GAN (LC-GAN)… ▽ More Intelligent vision is appealing in computer-assisted and robotic surgeries. Vision-based analysis with deep learning usually requires large labeled datasets, but manual data labeling is expensive and time-consuming in medical problems. We investigate a novel cross-domain strategy to reduce the need for manual data labeling by proposing an image-to-image translation model live-cadaver GAN (LC-GAN) based on generative adversarial networks (GANs). We consider a situation when a labeled cadaveric surgery dataset is available while the task is instrument segmentation on an unlabeled live surgery dataset. We train LC-GAN to learn the map**s between the cadaveric and live images. For live image segmentation, we first translate the live images to fake-cadaveric images with LC-GAN and then perform segmentation on the fake-cadaveric images with models trained on the real cadaveric dataset. The proposed method fully makes use of the labeled cadaveric dataset for live image segmentation without the need to label the live dataset. LC-GAN has two generators with different architectures that leverage the deep feature representation learned from the cadaveric image based segmentation task. Moreover, we propose the structural similarity loss and segmentation consistency loss to improve the semantic consistency during translation. Our model achieves better image-to-image translation and leads to improved segmentation performance in the proposed cross-domain segmentation task. △ Less

Submitted 13 August, 2020; v1 submitted 10 March, 2020; originally announced March 2020.

Comments: Accepted by 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

Freeze the Discriminator: a Simple Baseline for Fine-Tuning GANs

Authors: Sangwoo Mo, Minsu Cho, **woo Shin

Abstract: Generative adversarial networks (GANs) have shown outstanding performance on a wide range of problems in computer vision, graphics, and machine learning, but often require numerous training data and heavy computational resources. To tackle this issue, several methods introduce a transfer learning technique in GAN training. They, however, are either prone to overfitting or limited to learning small… ▽ More Generative adversarial networks (GANs) have shown outstanding performance on a wide range of problems in computer vision, graphics, and machine learning, but often require numerous training data and heavy computational resources. To tackle this issue, several methods introduce a transfer learning technique in GAN training. They, however, are either prone to overfitting or limited to learning small distribution shifts. In this paper, we show that simple fine-tuning of GANs with frozen lower layers of the discriminator performs surprisingly well. This simple baseline, FreezeD, significantly outperforms previous techniques used in both unconditional and conditional GANs. We demonstrate the consistent effect using StyleGAN and SNGAN-projection architectures on several datasets of Animal Face, Anime Face, Oxford Flower, CUB-200-2011, and Caltech-256 datasets. The code and results are available at https://github.com/sangwoomo/FreezeD. △ Less

Submitted 28 February, 2020; v1 submitted 25 February, 2020; originally announced February 2020.

Comments: Tech report; High resolution images are in https://github.com/sangwoomo/FreezeD

Towards Better Surgical Instrument Segmentation in Endoscopic Vision: Multi-Angle Feature Aggregation and Contour Supervision

Authors: Fangbo Qin, Shan Lin, Yangming Li, Randall A. Bly, Kris S. Moe, Blake Hannaford

Abstract: Accurate and real-time surgical instrument segmentation is important in the endoscopic vision of robot-assisted surgery, and significant challenges are posed by frequent instrument-tissue contacts and continuous change of observation perspective. For these challenging tasks more and more deep neural networks (DNN) models are designed in recent years. We are motivated to propose a general embeddabl… ▽ More Accurate and real-time surgical instrument segmentation is important in the endoscopic vision of robot-assisted surgery, and significant challenges are posed by frequent instrument-tissue contacts and continuous change of observation perspective. For these challenging tasks more and more deep neural networks (DNN) models are designed in recent years. We are motivated to propose a general embeddable approach to improve these current DNN segmentation models without increasing the model parameter number. Firstly, observing the limited rotation-invariance performance of DNN, we proposed the Multi-Angle Feature Aggregation (MAFA) method, leveraging active image rotation to gain richer visual cues and make the prediction more robust to instrument orientation changes. Secondly, in the end-to-end training stage, the auxiliary contour supervision is utilized to guide the model to learn the boundary awareness, so that the contour shape of segmentation mask is more precise. The proposed method is validated with ablation experiments on the novel Sinus-Surgery datasets collected from surgeons' operations, and is compared to the existing methods on a public dataset collected with a da Vinci Xi Robot. △ Less

Submitted 10 August, 2020; v1 submitted 25 February, 2020; originally announced February 2020.

Comments: Accepted by IEEE Robotics and Automation Letters

Lookahead: A Far-Sighted Alternative of Magnitude-based Pruning

Authors: Sejun Park, Jaeho Lee, Sangwoo Mo, **woo Shin

Abstract: Magnitude-based pruning is one of the simplest methods for pruning neural networks. Despite its simplicity, magnitude-based pruning and its variants demonstrated remarkable performances for pruning modern architectures. Based on the observation that magnitude-based pruning indeed minimizes the Frobenius distortion of a linear operator corresponding to a single layer, we develop a simple pruning me… ▽ More Magnitude-based pruning is one of the simplest methods for pruning neural networks. Despite its simplicity, magnitude-based pruning and its variants demonstrated remarkable performances for pruning modern architectures. Based on the observation that magnitude-based pruning indeed minimizes the Frobenius distortion of a linear operator corresponding to a single layer, we develop a simple pruning method, coined lookahead pruning, by extending the single layer optimization to a multi-layer optimization. Our experimental results demonstrate that the proposed method consistently outperforms magnitude-based pruning on various networks, including VGG and ResNet, particularly in the high-sparsity regime. See https://github.com/alinlab/lookahead_pruning for codes. △ Less

Submitted 12 February, 2020; originally announced February 2020.

Comments: ICLR 2020, camera ready

Accelerating Reinforcement Learning with Suboptimal Guidance

Authors: Eivind Bøhn, Signe Moe, Tor Arne Johansen

Abstract: Reinforcement Learning in domains with sparse rewards is a difficult problem, and a large part of the training process is often spent searching the state space in a more or less random fashion for any learning signals. For control problems, we often have some controller readily available which might be suboptimal but nevertheless solves the problem to some degree. This controller can be used to gu… ▽ More Reinforcement Learning in domains with sparse rewards is a difficult problem, and a large part of the training process is often spent searching the state space in a more or less random fashion for any learning signals. For control problems, we often have some controller readily available which might be suboptimal but nevertheless solves the problem to some degree. This controller can be used to guide the initial exploration phase of the learning controller towards reward yielding states, reducing the time before refinement of a viable policy can be initiated. In our work, the agent is guided through an auxiliary behaviour cloning loss which is made conditional on a Q-filter, i.e. it is only applied in situations where the critic deems the guiding controller to be better than the agent. The Q-filter provides a natural way to adjust the guidance throughout the training process, allowing the agent to exceed the guiding controller in a manner that is adaptive to the task at hand and the proficiency of the guiding controller. The contribution of this paper lies in identifying shortcomings in previously proposed implementations of the Q-filter concept, and in suggesting some ways these issues can be mitigated. These modifications are tested on the OpenAI Gym Fetch environments, showing clear improvements in adaptivity and yielding increased performance in all robotic environments tested. △ Less

Submitted 21 November, 2019; originally announced November 2019.

Comments: Submitted to IFAC 2020

Deep Reinforcement Learning Attitude Control of Fixed-Wing UAVs Using Proximal Policy Optimization

Authors: Eivind Bøhn, Erlend M. Coates, Signe Moe, Tor Arne Johansen

Abstract: Contemporary autopilot systems for unmanned aerial vehicles (UAVs) are far more limited in their flight envelope as compared to experienced human pilots, thereby restricting the conditions UAVs can operate in and the types of missions they can accomplish autonomously. This paper proposes a deep reinforcement learning (DRL) controller to handle the nonlinear attitude control problem, enabling exten… ▽ More Contemporary autopilot systems for unmanned aerial vehicles (UAVs) are far more limited in their flight envelope as compared to experienced human pilots, thereby restricting the conditions UAVs can operate in and the types of missions they can accomplish autonomously. This paper proposes a deep reinforcement learning (DRL) controller to handle the nonlinear attitude control problem, enabling extended flight envelopes for fixed-wing UAVs. A proof-of-concept controller using the proximal policy optimization (PPO) algorithm is developed, and is shown to be capable of stabilizing a fixed-wing UAV from a large set of initial conditions to reference roll, pitch and airspeed values. The training process is outlined and key factors for its progression rate are considered, with the most important factor found to be limiting the number of variables in the observation vector, and including values for several previous time steps for these variables. The trained reinforcement learning (RL) controller is compared to a proportional-integral-derivative (PID) controller, and is found to converge in more cases than the PID controller, with comparable performance. Furthermore, the RL controller is shown to generalize well to unseen disturbances in the form of wind and turbulence, even in severe disturbance conditions. △ Less

Submitted 13 November, 2019; originally announced November 2019.

Comments: 11 pages, 3 figures, 2019 International Conference on Unmanned Aircraft Systems (ICUAS)

Journal ref: In 2019 International Conference on Unmanned Aircraft Systems (ICUAS) (pp. 523-533). IEEE

Mining GOLD Samples for Conditional GANs

Authors: Sangwoo Mo, Chiheon Kim, Sungwoong Kim, Minsu Cho, **woo Shin

Abstract: Conditional generative adversarial networks (cGANs) have gained a considerable attention in recent years due to its class-wise controllability and superior quality for complex generation tasks. We introduce a simple yet effective approach to improving cGANs by measuring the discrepancy between the data distribution and the model distribution on given samples. The proposed measure, coined the gap o… ▽ More Conditional generative adversarial networks (cGANs) have gained a considerable attention in recent years due to its class-wise controllability and superior quality for complex generation tasks. We introduce a simple yet effective approach to improving cGANs by measuring the discrepancy between the data distribution and the model distribution on given samples. The proposed measure, coined the gap of log-densities (GOLD), provides an effective self-diagnosis for cGANs while being efficienty computed from the discriminator. We propose three applications of the GOLD: example re-weighting, rejection sampling, and active learning, which improve the training, inference, and data selection of cGANs, respectively. Our experimental results demonstrate that the proposed methods outperform corresponding baselines for all three applications on different image datasets. △ Less

Submitted 21 October, 2019; originally announced October 2019.

Comments: NeurIPS 2019

Electronic bandstructure of in-plane ferroelectric van der Waals $β'-In_{2}Se_{3}$

Authors: James L. Collins, Chutian Wang, Anton Tadich, Yuefeng Yin, Changxi Zheng, Jack Hellerstedt, Antonija Grubišić-Čabo, Shujie Tang, Sung-Kwan Mo, John Riley, Eric Huwald, Nikhil V. Medhekar, Michael S. Fuhrer, Mark T. Edmonds

Abstract: Layered indium selenides ($In_{2}Se_{3}$) have recently been discovered to host robust out-of-plane and in-plane ferroelectricity in the $α$ and $β$' phases, respectively. In this work, we utilise angle-resolved photoelectron spectroscopy to directly measure the electronic bandstructure of $β'-In_{2}Se_{3}$, and compare to hybrid density functional theory (DFT) calculations. In agreement with DFT,… ▽ More Layered indium selenides ($In_{2}Se_{3}$) have recently been discovered to host robust out-of-plane and in-plane ferroelectricity in the $α$ and $β$' phases, respectively. In this work, we utilise angle-resolved photoelectron spectroscopy to directly measure the electronic bandstructure of $β'-In_{2}Se_{3}$, and compare to hybrid density functional theory (DFT) calculations. In agreement with DFT, we find the band structure is highly two-dimensional, with negligible dispersion along the c-axis. Due to n-type do** we are able to observe the conduction band minima, and directly measure the minimum indirect (0.97 eV) and direct (1.46 eV) bandgaps. We find the Fermi surface in the conduction band is characterized by anisotropic electron pockets with sharp in-plane dispersion about the $\overline{M}$ points, yielding effective masses of 0.21 $m_{0}$ along $\overline{KM}$ and 0.33 $m_{0}$ along $\overline{ΓM}$. The measured band structure is well supported by hybrid density functional theory calculations. The highly two-dimensional (2D) bandstructure with moderate bandgap and small effective mass suggest that $β'-In_{2}Se_{3}$ is a potentially useful new van der Waals semiconductor. This together with its ferroelectricity makes it a viable material for high-mobility ferroelectric-photovoltaic devices, with applications in non-volatile memory switching and renewable energy technologies. △ Less

Submitted 17 February, 2020; v1 submitted 14 October, 2019; originally announced October 2019.

Comments: 19 pages, 4 + 1 figures; typos corrected;added references; updated figures & discussion to reflect changes in model

Magnetic Weyl Semimetal Phase in a Kagome Crystal

Authors: D. F. Liu, A. J. Liang, E. K. Liu, Q. N. Xu, Y. W. Li, C. Chen, D. Pei, W. J. Shi, S. K. Mo, P. Dudin, T. Kim, C. Cacho, G. Li, Y. Sun, L. X. Yang, Z. K. Liu, S. S. P. Parkin, C. Felser, Y. L. Chen

Abstract: Weyl semimetals are crystalline solids that host emergent relativistic Weyl fermions and have characteristic surface Fermi-arcs in their electronic structure. Weyl semimetals with broken time reversal symmetry are difficult to identify unambiguously. In this work, using angle-resolved photoemission spectroscopy, we visualized the electronic structure of the ferromagnetic crystal Co3Sn2S2 and disco… ▽ More Weyl semimetals are crystalline solids that host emergent relativistic Weyl fermions and have characteristic surface Fermi-arcs in their electronic structure. Weyl semimetals with broken time reversal symmetry are difficult to identify unambiguously. In this work, using angle-resolved photoemission spectroscopy, we visualized the electronic structure of the ferromagnetic crystal Co3Sn2S2 and discovered its characteristic surface Fermi-arcs and linear bulk band dispersions across the Weyl points. These results establish Co3Sn2S2 as a magnetic Weyl semimetal that may serve as a platform for realizing phenomena such as chiral magnetic effects, unusually large anomalous Hall effect and quantum anomalous Hall effect. △ Less

Submitted 20 September, 2019; originally announced September 2019.

Comments: 15 pages, 4 figures

Journal ref: Science 365,1282-1285 (2019)

Emergence of Quasiparticles in a Doped Mott Insulator

Authors: Yao Wang, Yu He, Krzysztof Wohlfeld, Makoto Hashimoto, Edwin W. Huang, Donghui Lu, Sung-Kwan Mo, Seiki Komiya, Chun**g Jia, Brian Moritz, Zhi-Xun Shen, Thomas P. Devereaux

Abstract: How a Mott insulator develops into a weakly coupled metal upon do** is a central question to understanding various emergent correlated phenomena. To analyze this evolution and its connection to the high-$T_c$ cuprates, we study the single-particle spectrum for the doped Hubbard model using cluster perturbation theory on superclusters. Starting from extremely low do**, we identify a heavily ren… ▽ More How a Mott insulator develops into a weakly coupled metal upon do** is a central question to understanding various emergent correlated phenomena. To analyze this evolution and its connection to the high-$T_c$ cuprates, we study the single-particle spectrum for the doped Hubbard model using cluster perturbation theory on superclusters. Starting from extremely low do**, we identify a heavily renormalized quasiparticle dispersion that immediately develops across the Fermi level, and a weakening polaronic side band at higher binding energy. The quasiparticle spectral weight roughly grows at twice the rate of do** in the low do** regime, but this rate is halved at optimal do**. In the heavily doped regime, we find both strong electron-hole asymmetry and a persistent presence of Mott spectral features. Finally, we discuss the applicability of the single-band Hubbard model to describe the evolution of nodal spectra measured by angle-resolved photoemission spectroscopy (ARPES) on the single-layer cuprate La$_{2-x}$Sr$_x$CuO$_4$ ($0 \le x \le 0.15$). This work benchmarks the predictive power of the Hubbard model for electronic properties of high-$T_c$ cuprates. △ Less

Submitted 15 November, 2020; v1 submitted 19 September, 2019; originally announced September 2019.

Comments: 7 pages, 5 figures

Journal ref: Commun. Phys. 3, 210 (2020)

Integration of adversarial autoencoders with residual dense convolutional networks for estimation of non-Gaussian hydraulic conductivities

Authors: Shaoxing Mo, Nicholas Zabaras, Xiaoqing Shi, Jichun Wu

Abstract: Inverse modeling for the estimation of non-Gaussian hydraulic conductivity fields in subsurface flow and solute transport models remains a challenging problem. This is mainly due to the non-Gaussian property, the non-linear physics, and the fact that many repeated evaluations of the forward model are often required. In this study, we develop a convolutional adversarial autoencoder (CAAE) to parame… ▽ More Inverse modeling for the estimation of non-Gaussian hydraulic conductivity fields in subsurface flow and solute transport models remains a challenging problem. This is mainly due to the non-Gaussian property, the non-linear physics, and the fact that many repeated evaluations of the forward model are often required. In this study, we develop a convolutional adversarial autoencoder (CAAE) to parameterize non-Gaussian conductivity fields with heterogeneous conductivity within each facies using a low-dimensional latent representation. In addition, a deep residual dense convolutional network (DRDCN) is proposed for surrogate modeling of forward models with high-dimensional and highly-complex map**s. The two networks are both based on a multilevel residual learning architecture called residual-in-residual dense block. The multilevel residual learning strategy and the dense connection structure ease the training of deep networks, enabling us to efficiently build deeper networks that have an essentially increased capacity for approximating map**s of very high-complexity. The CCAE and DRDCN networks are incorporated into an iterative ensemble smoother to formulate an inversion framework. The numerical experiments performed using 2-D and 3-D solute transport models illustrate the performance of the integrated method. The obtained results indicate that the CAAE is a robust parameterization method for non-Gaussian conductivity fields with different heterogeneity patterns. The DRDCN is able to obtain accurate approximations of the forward models with high-dimensional and highly-complex map**s using relatively limited training data. The CAAE and DRDCN methods together significantly reduce the computation time required to achieve accurate inversion results. △ Less

Submitted 13 January, 2020; v1 submitted 26 June, 2019; originally announced June 2019.

Extremely large magnetoresistance and compensated Fermi surfaces in the antiferromagnetic semimetal YbAs

Authors: W. Xie, Y. Wu, F. Du, A. Wang, H. Su, Y. Chen, Z. Y. Nie, S. -K. Mo, M. Smidman, C. Cao, Y. Liu, T. Takabatake, H. Q. Yuan

Abstract: A number of rare-earth monopnictides have topologically non-trivial band structures together with magnetism and strong electronic correlations. In order to examine whether the antiferromagnetic (AFM) semimetal YbAs ($T\rm_N$ = 0.5 K) exhibits such a scenario, we have grown high-quality single crystals using a flux method, and characterized the magnetic properties and electronic structure using spe… ▽ More A number of rare-earth monopnictides have topologically non-trivial band structures together with magnetism and strong electronic correlations. In order to examine whether the antiferromagnetic (AFM) semimetal YbAs ($T\rm_N$ = 0.5 K) exhibits such a scenario, we have grown high-quality single crystals using a flux method, and characterized the magnetic properties and electronic structure using specific heat, magnetotransport and angle-resolved photoemission spectroscopy (ARPES) measurements, together with density functional theory (DFT) calculations. Both ARPES and DFT calculations find no evidence for band inversions in YbAs, indicating a topologically trivial electronic structure. From low-temperature magnetotransport measurements, we map the field-temperature phase diagram, where we find the presence of a field stabilized phase distinct from the AFM phase at low temperatures. An extremely large magnetoresistance (XMR) for both YbAs and the nonmagnetic counterpart LuAs, is also observed, which can consistently be accounted for by the presence of electron-hole compensation. Moreover, an angle-dependent study of the Shubnikov-de Haas effect oscillations reveals very similar Fermi surfaces between YbAs and LuAs, with light effective masses down to at least 0.5 K, indicating that the Yb-$4f$ electrons are well localized, and do not contribute to the Fermi surface. However, the influence of the localized Yb-$4f$ electrons on the magnetotransport of YbAs can be discerned from the distinct temperature dependence of the XMR compared to that of LuAs, which we attribute to the influence of short-ranged spin correlations that appear well above $T\rm_N$. △ Less

Submitted 20 February, 2020; v1 submitted 26 June, 2019; originally announced June 2019.

Comments: 12 pages, 10 figures

Journal ref: Phys. Rev. B 101, 085132 (2020)

Visualizing Exotic Orbital Texture in the Single-Layer Mott Insulator 1T-TaSe2

Authors: Yi Chen, Wei Ruan, Meng Wu, Shujie Tang, Hye** Ryu, Hsin-Zon Tsai, Ryan Lee, Salman Kahn, Franklin Liou, Caihong Jia, Oliver R. Albertini, Hongyu Xiong, Tao Jia, Zhi Liu, Jonathan A. Sobota, Amy Y. Liu, Joel E. Moore, Zhi-Xun Shen, Steven G. Louie, Sung-Kwan Mo, Michael F. Crommie

Abstract: Mott insulating behavior is induced by strong electron correlation and can lead to exotic states of matter such as unconventional superconductivity and quantum spin liquids. Recent advances in van der Waals material synthesis enable the exploration of novel Mott systems in the two-dimensional limit. Here we report characterization of the local electronic properties of single- and few-layer 1T-TaSe… ▽ More Mott insulating behavior is induced by strong electron correlation and can lead to exotic states of matter such as unconventional superconductivity and quantum spin liquids. Recent advances in van der Waals material synthesis enable the exploration of novel Mott systems in the two-dimensional limit. Here we report characterization of the local electronic properties of single- and few-layer 1T-TaSe2 via spatial- and momentum-resolved spectroscopy involving scanning tunneling microscopy and angle-resolved photoemission. Our combined experimental and theoretical study indicates that electron correlation induces a robust Mott insulator state in single-layer 1T-TaSe2 that is accompanied by novel orbital texture. Inclusion of interlayer coupling weakens the insulating phase in 1T-TaSe2, as seen by strong reduction of its energy gap and quenching of its correlation-driven orbital texture in bilayer and trilayer 1T-TaSe2. Our results establish single-layer 1T-TaSe2 as a useful new platform for investigating strong correlation physics in two dimensions. △ Less

Submitted 16 May, 2019; v1 submitted 24 April, 2019; originally announced April 2019.

Journal ref: Nature Physics 16, 218-224 (2020)

InstaGAN: Instance-aware Image-to-Image Translation

Authors: Sangwoo Mo, Minsu Cho, **woo Shin

Abstract: Unsupervised image-to-image translation has gained considerable attention due to the recent impressive progress based on generative adversarial networks (GANs). However, previous methods often fail in challenging cases, in particular, when an image has multiple target instances and a translation task involves significant changes in shape, e.g., translating pants to skirts in fashion images. To tac… ▽ More Unsupervised image-to-image translation has gained considerable attention due to the recent impressive progress based on generative adversarial networks (GANs). However, previous methods often fail in challenging cases, in particular, when an image has multiple target instances and a translation task involves significant changes in shape, e.g., translating pants to skirts in fashion images. To tackle the issues, we propose a novel method, coined instance-aware GAN (InstaGAN), that incorporates the instance information (e.g., object segmentation masks) and improves multi-instance transfiguration. The proposed method translates both an image and the corresponding set of instance attributes while maintaining the permutation invariance property of the instances. To this end, we introduce a context preserving loss that encourages the network to learn the identity function outside of target instances. We also propose a sequential mini-batch inference/training technique that handles multiple instances with a limited GPU memory and enhances the network to generalize better for multiple instances. Our comparative evaluation demonstrates the effectiveness of the proposed method on different image datasets, in particular, in the aforementioned challenging cases. Code and results are available in https://github.com/sangwoomo/instagan △ Less

Submitted 2 January, 2019; v1 submitted 27 December, 2018; originally announced December 2018.

Comments: Accepted to ICLR 2019. High resolution images are available in https://github.com/sangwoomo/instagan

Deep autoregressive neural networks for high-dimensional inverse problems in groundwater contaminant source identification

Authors: Shaoxing Mo, Nicholas Zabaras, Xiaoqing Shi, Jichun Wu

Abstract: Identification of a groundwater contaminant source simultaneously with the hydraulic conductivity in highly-heterogeneous media often results in a high-dimensional inverse problem. In this study, a deep autoregressive neural network-based surrogate method is developed for the forward model to allow us to solve efficiently such high-dimensional inverse problems. The surrogate is trained using limit… ▽ More Identification of a groundwater contaminant source simultaneously with the hydraulic conductivity in highly-heterogeneous media often results in a high-dimensional inverse problem. In this study, a deep autoregressive neural network-based surrogate method is developed for the forward model to allow us to solve efficiently such high-dimensional inverse problems. The surrogate is trained using limited evaluations of the forward model. Since the relationship between the time-varying inputs and outputs of the forward transport model is complex, we propose an autoregressive strategy, which treats the output at the previous time step as input to the network for predicting the output at the current time step. We employ a dense convolutional encoder-decoder network architecture in which the high-dimensional input and output fields of the model are treated as images to leverage the robust capability of convolutional networks in image-like data processing. An iterative local updating ensemble smoother (ILUES) algorithm is used as the inversion framework. The proposed method is evaluated using a synthetic contaminant source identification problem with 686 uncertain input parameters. Results indicate that, with relatively limited training data, the deep autoregressive neural network consisting of 27 convolutional layers is capable of providing an accurate approximation for the high-dimensional model input-output relationship. The autoregressive strategy substantially improves the network's accuracy and computational efficiency. The application of the surrogate-based ILUES in solving the inverse problem shows that it can achieve accurate inversion results and predictive uncertainty estimates. △ Less

Submitted 21 December, 2018; originally announced December 2018.

Comments: 30 pages, 21 figures, submitted to Water Resources Research

Unique gap structure and symmetry of the charge density wave in single-layer VSe$_2$

Authors: P. Chen, W. -W. Pai, Y. -H. Chan, V. Madhavan, M. Y. Chou, S. -K. Mo, A. -V. Fedorov, T. -C. Chiang

Abstract: Single layers of transition metal dichalcogenides (TMDCs) are excellent candidates for electronic applications beyond the graphene platform; many of them exhibit novel properties including charge density waves (CDWs) and magnetic ordering. CDWs in these single layers are generally a planar projection of the corresponding bulk CDWs because of the quasi-two-dimensional nature of TMDCs; a different C… ▽ More Single layers of transition metal dichalcogenides (TMDCs) are excellent candidates for electronic applications beyond the graphene platform; many of them exhibit novel properties including charge density waves (CDWs) and magnetic ordering. CDWs in these single layers are generally a planar projection of the corresponding bulk CDWs because of the quasi-two-dimensional nature of TMDCs; a different CDW symmetry is unexpected. We report herein the successful creation of pristine single-layer VSe$_2$, which shows a ($\sqrt7 \times \sqrt3$) CDW in contrast to the (4 $\times$ 4) CDW for the layers in bulk VSe$_2$. Angle-resolved photoemission spectroscopy (ARPES) from the single layer shows a sizable ($\sqrt7 \times \sqrt3$) CDW gap of $\sim$100 meV at the zone boundary, a 220 K CDW transition temperature twice the bulk value, and no ferromagnetic exchange splitting as predicted by theory. This robust CDW with an exotic broken symmetry as the ground state is explained via a first-principles analysis. The results illustrate a unique CDW phenomenon in the two-dimensional limit. △ Less

Submitted 14 November, 2018; originally announced November 2018.

Journal ref: Phys. Rev. Lett. 121, 196402 (2018)

Metallic surface states in a correlated d-electron topological Kondo insulator candidate FeSb2

Authors: Ke-Jun Xu, Su-Di Chen, Yu He, Junfeng He, Shujie Tang, Chun**g Jia, Eric Yue Ma, Sung-Kwan Mo, Dong-Hui Lu, Makoto Hashimoto, Thomas P. Devereaux, Zhi-Xun Shen

Abstract: The resistance of a conventional insulator diverges as temperature approaches zero. The peculiar low temperature resistivity saturation in the 4f Kondo insulator (KI) SmB6 has spurred proposals of a correlation-driven topological Kondo insulator (TKI) with exotic ground states. However, the scarcity of model TKI material families leaves difficulties in disentangling key ingredients from irrelevant… ▽ More The resistance of a conventional insulator diverges as temperature approaches zero. The peculiar low temperature resistivity saturation in the 4f Kondo insulator (KI) SmB6 has spurred proposals of a correlation-driven topological Kondo insulator (TKI) with exotic ground states. However, the scarcity of model TKI material families leaves difficulties in disentangling key ingredients from irrelevant details. Here we use angle-resolved photoemission spectroscopy (ARPES) to study FeSb2, a correlated d-electron KI candidate that also exhibits a low temperature resistivity saturation. On the (010) surface, we find a rich assemblage of metallic states with two-dimensional dispersion. Measurements of the bulk band structure reveal band renormalization, a large temperature-dependent band shift, and flat spectral features along certain high symmetry directions, providing spectroscopic evidence for strong correlations. Our observations suggest that exotic insulating states resembling those in SmB6 and YbB12 may also exist in systems with d instead of f electrons. △ Less

Submitted 13 May, 2020; v1 submitted 6 November, 2018; originally announced November 2018.