Showing 1–36 of 36 results for author: Choe, J H

Addressing Relative Pose Impact on UWB Localization: Dataset Introduction and Analysis

Authors: Jun Hyeok Choe, Inwook Shim

Abstract: UWB has recently gained new attention as an auxiliary sensor in the field of robot localization due to its compactness and ease of distance measurement. Consequently, various UWB-related localization and dataset research have increased. Despite this broad interest, there is a lack of UWB datasets that thoroughly analyze the performance of UWB ranging measurement. To address this issue, our paper i… ▽ More UWB has recently gained new attention as an auxiliary sensor in the field of robot localization due to its compactness and ease of distance measurement. Consequently, various UWB-related localization and dataset research have increased. Despite this broad interest, there is a lack of UWB datasets that thoroughly analyze the performance of UWB ranging measurement. To address this issue, our paper introduces a UWB dataset that examines UWB relative pose factors affecting ranging measurement. To the best of our knowledge, our dataset is the first to analyze these factors while rigorously providing precise ground-truth UWB poses. The dataset is accessible at https://github.com/cjhhalla/RCV_uwb_dataset . △ Less

Submitted 4 July, 2024; originally announced July 2024.

Comments: 4 pages

Language-Image Models with 3D Understanding

Authors: Jang Hyun Cho, Boris Ivanovic, Yulong Cao, Edward Schmerling, Yue Wang, Xinshuo Weng, Boyi Li, Yurong You, Philipp Krähenbühl, Yan Wang, Marco Pavone

Abstract: Multi-modal large language models (MLLMs) have shown incredible capabilities in a variety of 2D vision and language tasks. We extend MLLMs' perceptual capabilities to ground and reason about images in 3-dimensional space. To that end, we first develop a large-scale pre-training dataset for 2D and 3D called LV3D by combining multiple existing 2D and 3D recognition datasets under a common task formu… ▽ More Multi-modal large language models (MLLMs) have shown incredible capabilities in a variety of 2D vision and language tasks. We extend MLLMs' perceptual capabilities to ground and reason about images in 3-dimensional space. To that end, we first develop a large-scale pre-training dataset for 2D and 3D called LV3D by combining multiple existing 2D and 3D recognition datasets under a common task formulation: as multi-turn question-answering. Next, we introduce a new MLLM named Cube-LLM and pre-train it on LV3D. We show that pure data scaling makes a strong 3D perception capability without 3D specific architectural design or training objective. Cube-LLM exhibits intriguing properties similar to LLMs: (1) Cube-LLM can apply chain-of-thought prompting to improve 3D understanding from 2D context information. (2) Cube-LLM can follow complex and diverse instructions and adapt to versatile input and output formats. (3) Cube-LLM can be visually prompted such as 2D box or a set of candidate 3D boxes from specialists. Our experiments on outdoor benchmarks demonstrate that Cube-LLM significantly outperforms existing baselines by 21.3 points of AP-BEV on the Talk2Car dataset for 3D grounded reasoning and 17.7 points on the DriveLM dataset for complex reasoning about driving scenarios, respectively. Cube-LLM also shows competitive results in general MLLM benchmarks such as refCOCO for 2D grounding with (87.0) average score, as well as visual question answering benchmarks such as VQAv2, GQA, SQA, POPE, etc. for complex reasoning. Our project is available at https://janghyuncho.github.io/Cube-LLM. △ Less

Submitted 6 May, 2024; originally announced May 2024.

Comments: Project page: https://janghyuncho.github.io/Cube-LLM

Event-based Product Carousel Recommendation with Query-Click Graph

Authors: Luyi Ma, Nimesh Sinha, Parth Vajge, Jason HD Cho, Sushant Kumar, Kannan Achan

Abstract: Many current recommender systems mainly focus on the product-to-product recommendations and user-to-product recommendations even during the time of events rather than modeling the typical recommendations for the target event (e.g., festivals, seasonal activities, or social activities) without addressing the multiple aspects of the shop** demands for the target event. Product recommendations for… ▽ More Many current recommender systems mainly focus on the product-to-product recommendations and user-to-product recommendations even during the time of events rather than modeling the typical recommendations for the target event (e.g., festivals, seasonal activities, or social activities) without addressing the multiple aspects of the shop** demands for the target event. Product recommendations for the multiple aspects of the target event are usually generated by human curators who manually identify the aspects and select a list of aspect-related products (i.e., product carousel) for each aspect as recommendations. However, building a recommender system with machine learning is non-trivial due to the lack of both the ground truth of event-related aspects and the aspect-related products. To fill this gap, we define the novel problem as the event-based product carousel recommendations in e-commerce and propose an effective recommender system based on the query-click bipartite graph. We apply the iterative clustering algorithm over the query-click bipartite graph and infer the event-related aspects by the clusters of queries. The aspect-related recommendations are powered by the click-through rate of products regarding each aspect. We show through experiments that this approach effectively mines product carousels for the target event. △ Less

Submitted 5 February, 2024; originally announced February 2024.

Comments: 7 pages, 2 figures, 2021 IEEE International Conference on Big Data (Big Data)

Spreading of Low-viscosity Ink Filaments Driven by Bath Viscoelasticity in Embedded Printing

Authors: Jae Hyung Cho, Emilie Dressaire

Abstract: Inks deposited in conventional direct ink writing need to be able to support their own weight and that of the upper layers with minimal deformation to preserve the structural integrity of the three-dimensional (3D) printed parts. This constraint limits the range of usable inks to high-viscosity materials. Embedded printing enables the use of much softer inks by depositing the materials in a bath o… ▽ More Inks deposited in conventional direct ink writing need to be able to support their own weight and that of the upper layers with minimal deformation to preserve the structural integrity of the three-dimensional (3D) printed parts. This constraint limits the range of usable inks to high-viscosity materials. Embedded printing enables the use of much softer inks by depositing the materials in a bath of another fluid that provides external support, thus diversifying the types of 3D printable structures. The interactions between the ink and bath fluids, however, give rise to a unique type of defect: spreading of the dispensed ink behind the moving nozzle. By printing horizontal threads made of dyed water in baths of Carbopol suspensions, we demonstrate that the spreading can be attributed to the pressure field generated in the viscous bath by the relative motion of the nozzle. As the pressure gradient increases with the viscosity of the bath fluid while the viscosity of the ink resists the flow, a larger bath-to-ink viscosity ratio results in more spreading for low-concentration Carbopol baths. For high-concentration, yield-stress-fluid baths, we find that the steady-state viscosity alone cannot account for the spreading, as the elastic stress becomes comparable to the viscous stress and the bath fluid around the dispensed ink undergoes fluidization and resolidification. By parameterizing the transient rheology of the high-concentration Carbopol suspensions using a simple viscoelastic model, we suggest that the ink spreading is exacerbated by the elasticity but is mitigated by the yield stress as long as the yield stress is low enough to allow steady injection of the ink. These results help illuminate the link between the bath rheology and the printing quality in embedded 3D printing. △ Less

Submitted 17 January, 2024; originally announced January 2024.

Language-conditioned Detection Transformer

Authors: Jang Hyun Cho, Philipp Krähenbühl

Abstract: We present a new open-vocabulary detection framework. Our framework uses both image-level labels and detailed detection annotations when available. Our framework proceeds in three steps. We first train a language-conditioned object detector on fully-supervised detection data. This detector gets to see the presence or absence of ground truth classes during training, and conditions prediction on the… ▽ More We present a new open-vocabulary detection framework. Our framework uses both image-level labels and detailed detection annotations when available. Our framework proceeds in three steps. We first train a language-conditioned object detector on fully-supervised detection data. This detector gets to see the presence or absence of ground truth classes during training, and conditions prediction on the set of present classes. We use this detector to pseudo-label images with image-level labels. Our detector provides much more accurate pseudo-labels than prior approaches with its conditioning mechanism. Finally, we train an unconditioned open-vocabulary detector on the pseudo-annotated images. The resulting detector, named DECOLA, shows strong zero-shot performance in open-vocabulary LVIS benchmark as well as direct zero-shot transfer benchmarks on LVIS, COCO, Object365, and OpenImages. DECOLA outperforms the prior arts by 17.1 AP-rare and 9.4 mAP on zero-shot LVIS benchmark. DECOLA achieves state-of-the-art results in various model sizes, architectures, and datasets by only training on open-sourced data and academic-scale computing. Code is available at https://github.com/janghyuncho/DECOLA. △ Less

Submitted 29 November, 2023; originally announced November 2023.

Comments: Code is at https://github.com/janghyuncho/DECOLA

Knowledge Graph Completion Models are Few-shot Learners: An Empirical Study of Relation Labeling in E-commerce with LLMs

Authors: Jiao Chen, Luyi Ma, Xiaohan Li, Nikhil Thakurdesai, Jianpeng Xu, Jason H. D. Cho, Kaushiki Nag, Evren Korpeoglu, Sushant Kumar, Kannan Achan

Abstract: Knowledge Graphs (KGs) play a crucial role in enhancing e-commerce system performance by providing structured information about entities and their relationships, such as complementary or substitutable relations between products or product types, which can be utilized in recommender systems. However, relation labeling in KGs remains a challenging task due to the dynamic nature of e-commerce domains… ▽ More Knowledge Graphs (KGs) play a crucial role in enhancing e-commerce system performance by providing structured information about entities and their relationships, such as complementary or substitutable relations between products or product types, which can be utilized in recommender systems. However, relation labeling in KGs remains a challenging task due to the dynamic nature of e-commerce domains and the associated cost of human labor. Recently, breakthroughs in Large Language Models (LLMs) have shown surprising results in numerous natural language processing tasks. In this paper, we conduct an empirical study of LLMs for relation labeling in e-commerce KGs, investigating their powerful learning capabilities in natural language and effectiveness in predicting relations between product types with limited labeled data. We evaluate various LLMs, including PaLM and GPT-3.5, on benchmark datasets, demonstrating their ability to achieve competitive performance compared to humans on relation labeling tasks using just 1 to 5 labeled examples per relation. Additionally, we experiment with different prompt engineering techniques to examine their impact on model performance. Our results show that LLMs significantly outperform existing KG completion models in relation labeling for e-commerce KGs and exhibit performance strong enough to replace human labeling. △ Less

Submitted 16 May, 2023; originally announced May 2023.

Long-tail Detection with Effective Class-Margins

Authors: Jang Hyun Cho, Philipp Krähenbühl

Abstract: Large-scale object detection and instance segmentation face a severe data imbalance. The finer-grained object classes become, the less frequent they appear in our datasets. However, at test-time, we expect a detector that performs well for all classes and not just the most frequent ones. In this paper, we provide a theoretical understanding of the long-trail detection problem. We show how the comm… ▽ More Large-scale object detection and instance segmentation face a severe data imbalance. The finer-grained object classes become, the less frequent they appear in our datasets. However, at test-time, we expect a detector that performs well for all classes and not just the most frequent ones. In this paper, we provide a theoretical understanding of the long-trail detection problem. We show how the commonly used mean average precision evaluation metric on an unknown test set is bound by a margin-based binary classification error on a long-tailed object detection training set. We optimize margin-based binary classification error with a novel surrogate objective called \textbf{Effective Class-Margin Loss} (ECM). The ECM loss is simple, theoretically well-motivated, and outperforms other heuristic counterparts on LVIS v1 benchmark over a wide range of architecture and detectors. Code is available at \url{https://github.com/janghyuncho/ECM-Loss}. △ Less

Submitted 23 January, 2023; originally announced January 2023.

Comments: ECCV 2022 Oral. Code is available at https://github.com/janghyuncho/ECM-Loss

NMS Strikes Back

Authors: Jeffrey Ouyang-Zhang, Jang Hyun Cho, Xingyi Zhou, Philipp Krähenbühl

Abstract: Detection Transformer (DETR) directly transforms queries to unique objects by using one-to-one bipartite matching during training and enables end-to-end object detection. Recently, these models have surpassed traditional detectors on COCO with undeniable elegance. However, they differ from traditional detectors in multiple designs, including model architecture and training schedules, and thus the… ▽ More Detection Transformer (DETR) directly transforms queries to unique objects by using one-to-one bipartite matching during training and enables end-to-end object detection. Recently, these models have surpassed traditional detectors on COCO with undeniable elegance. However, they differ from traditional detectors in multiple designs, including model architecture and training schedules, and thus the effectiveness of one-to-one matching is not fully understood. In this work, we conduct a strict comparison between the one-to-one Hungarian matching in DETRs and the one-to-many label assignments in traditional detectors with non-maximum supervision (NMS). Surprisingly, we observe one-to-many assignments with NMS consistently outperform standard one-to-one matching under the same setting, with a significant gain of up to 2.5 mAP. Our detector that trains Deformable-DETR with traditional IoU-based label assignment achieved 50.2 COCO mAP within 12 epochs (1x schedule) with ResNet50 backbone, outperforming all existing traditional or transformer-based detectors in this setting. On multiple datasets, schedules, and architectures, we consistently show bipartite matching is unnecessary for performant detection transformers. Furthermore, we attribute the success of detection transformers to their expressive transformer architecture. Code is available at https://github.com/jozhang97/DETA. △ Less

Submitted 12 December, 2022; originally announced December 2022.

Comments: Code is available at https://github.com/jozhang97/DETA

NEAT: A Label Noise-resistant Complementary Item Recommender System with Trustworthy Evaluation

Authors: Luyi Ma, Jianpeng Xu, Jason H. D. Cho, Evren Korpeoglu, Sushant Kumar, Kannan Achan

Abstract: The complementary item recommender system (CIRS) recommends the complementary items for a given query item. Existing CIRS models consider the item co-purchase signal as a proxy of the complementary relationship due to the lack of human-curated labels from the huge transaction records. These methods represent items in a complementary embedding space and model the complementary relationship as a poi… ▽ More The complementary item recommender system (CIRS) recommends the complementary items for a given query item. Existing CIRS models consider the item co-purchase signal as a proxy of the complementary relationship due to the lack of human-curated labels from the huge transaction records. These methods represent items in a complementary embedding space and model the complementary relationship as a point estimation of the similarity between items vectors. However, co-purchased items are not necessarily complementary to each other. For example, customers may frequently purchase bananas and bottled water within the same transaction, but these two items are not complementary. Hence, using co-purchase signals directly as labels will aggravate the model performance. On the other hand, the model evaluation will not be trustworthy if the labels for evaluation are not reflecting the true complementary relatedness. To address the above challenges from noisy labeling of the copurchase data, we model the co-purchases of two items as a Gaussian distribution, where the mean denotes the co-purchases from the complementary relatedness, and covariance denotes the co-purchases from the noise. To do so, we represent each item as a Gaussian embedding and parameterize the Gaussian distribution of co-purchases by the means and covariances from item Gaussian embedding. To reduce the impact of the noisy labels during evaluation, we propose an independence test-based method to generate a trustworthy label set with certain confidence. Our extensive experiments on both the publicly available dataset and the large-scale real-world dataset justify the effectiveness of our proposed model in complementary item recommendations compared with the state-of-the-art models. △ Less

Submitted 11 February, 2022; originally announced February 2022.

Comments: 11 pages, 4 figures; Published in: 2021 IEEE International Conference on Big Data (Big Data)

Yield precursor in primary creep of colloidal gels

Authors: Jae Hyung Cho, Irmgard Bischofberger

Abstract: Predicting the time-dependent yielding of colloidal gels under constant stress enables control of their mechanical stability and transport. Using rotational rheometry, we show that the shear rate of colloidal gels during an early stage of deformation known as primary creep can forecast an eventual yielding. Irrespective of whether the gel strain-softens or strain-hardens, the shear rate before fai… ▽ More Predicting the time-dependent yielding of colloidal gels under constant stress enables control of their mechanical stability and transport. Using rotational rheometry, we show that the shear rate of colloidal gels during an early stage of deformation known as primary creep can forecast an eventual yielding. Irrespective of whether the gel strain-softens or strain-hardens, the shear rate before failure exhibits a characteristic power-law decrease as a function of time, distinct from the linear viscoelastic response. We model this early-stage behavior as a series of uncorrelated local plastic events that are thermally activated, which illuminates the exponential dependence of the yield time on the applied stress. This precursor to yield in the macroscopic shear rate provides a convenient tool to identify the fate of a gel well in advance of the actual yielding. △ Less

Submitted 8 December, 2021; originally announced December 2021.

Lasting effects of discontinuous shear thickening in cornstarch suspensions upon flow cessation

Authors: Jae Hyung Cho, Andrew H. Griese, Ivo R. Peters, Irmgard Bischofberger

Abstract: Dense suspensions that exhibit discontinuous shear thickening (DST) undergo complex stress relaxation when the flow abruptly stops. Using rotational rheometry, we study the two-step relaxation of aqueous cornstarch suspensions out of the DST state upon flow cessation and show that the DST fluid retains the memory of its shear-thickening state until the shear stress reaches a constant value at late… ▽ More Dense suspensions that exhibit discontinuous shear thickening (DST) undergo complex stress relaxation when the flow abruptly stops. Using rotational rheometry, we study the two-step relaxation of aqueous cornstarch suspensions out of the DST state upon flow cessation and show that the DST fluid retains the memory of its shear-thickening state until the shear stress reaches a constant value at late times. We find that this residual stress at the end of the relaxation increases with the steady-state viscosity before the cessation. Furthermore, the timescales that characterize the two-step exponential decay of the shear stress exhibit near linear dependence on the steady-state viscosity. Within the current framework that ascribes DST to the breakdown of hydrodynamic lubrication layers leading to interparticle frictional contacts, the lasting effects of the steady-state viscosity suggest that the memory of frictional contacts persists until the end of the relaxation, despite the presence of repulsive forces between the particles. These results indicate that complete, spontaneous relaxation of the system out of DST is stalled by the partial retention of the frictional force chains, which may be caused by the stationary boundaries and the adhesion between cornstarch particles. △ Less

Submitted 2 November, 2021; originally announced November 2021.

PiCIE: Unsupervised Semantic Segmentation using Invariance and Equivariance in Clustering

Authors: Jang Hyun Cho, Utkarsh Mall, Kavita Bala, Bharath Hariharan

Abstract: We present a new framework for semantic segmentation without annotations via clustering. Off-the-shelf clustering methods are limited to curated, single-label, and object-centric images yet real-world data are dominantly uncurated, multi-label, and scene-centric. We extend clustering from images to pixels and assign separate cluster membership to different instances within each image. However, sol… ▽ More We present a new framework for semantic segmentation without annotations via clustering. Off-the-shelf clustering methods are limited to curated, single-label, and object-centric images yet real-world data are dominantly uncurated, multi-label, and scene-centric. We extend clustering from images to pixels and assign separate cluster membership to different instances within each image. However, solely relying on pixel-wise feature similarity fails to learn high-level semantic concepts and overfits to low-level visual cues. We propose a method to incorporate geometric consistency as an inductive bias to learn invariance and equivariance for photometric and geometric variations. With our novel learning objective, our framework can learn high-level semantic concepts. Our method, PiCIE (Pixel-level feature Clustering using Invariance and Equivariance), is the first method capable of segmenting both things and stuff categories without any hyperparameter tuning or task-specific pre-processing. Our method largely outperforms existing baselines on COCO and Cityscapes with +17.5 Acc. and +4.5 mIoU. We show that PiCIE gives a better initialization for standard supervised training. The code is available at https://github.com/janghyuncho/PiCIE. △ Less

Submitted 29 March, 2021; originally announced March 2021.

Comments: CVPR 2021

Two Modes of Cluster Dynamics Govern the Viscoelasticity of Colloidal Gels

Authors: Jae Hyung Cho, Irmgard Bischofberger

Abstract: Colloidal gels formed by strongly attractive particles at low particle volume fractions are composed of space-spanning networks of uniformly sized clusters. We study the thermal fluctuations of the clusters using differential dynamic microscopy by decomposing them into two modes of dynamics, and link them to the macroscopic viscoelasticity via rheometry. The first mode, dominant at early times, re… ▽ More Colloidal gels formed by strongly attractive particles at low particle volume fractions are composed of space-spanning networks of uniformly sized clusters. We study the thermal fluctuations of the clusters using differential dynamic microscopy by decomposing them into two modes of dynamics, and link them to the macroscopic viscoelasticity via rheometry. The first mode, dominant at early times, represents the localized, elastic fluctuations of individual clusters. The second mode, pronounced at late times, reflects the collective, viscoelastic dynamics facilitated by the connectivity of the clusters. By mixing two types of particles of distinct attraction strengths in different proportions, we control the transition time at which the collective mode starts to dominate, and hence tune the frequency dependence of the linear viscoelastic moduli of the binary gels. △ Less

Submitted 2 March, 2021; originally announced March 2021.

Journal ref: Phys. Rev. E 103, 032609 (2021)

Complete solution over $\GF{p^n}$ of the equation $X^{p^k+1}+X+a=0$

Authors: Kwang Ho Kim, Jong Hyok Choe, Sihem Mesnager

Abstract: The problem of solving explicitly the equation $P_a(X):=X^{q+1}+X+a=0$ over the finite field $\GF{Q}$, where $Q=p^n$, $q=p^k$ and $p$ is a prime, arises in many different contexts including finite geometry, the inverse Galois problem \cite{ACZ2000}, the construction of difference sets with Singer parameters \cite{DD2004}, determining cross-correlation between $m$-sequences \cite{DOBBERTIN2006} and… ▽ More The problem of solving explicitly the equation $P_a(X):=X^{q+1}+X+a=0$ over the finite field $\GF{Q}$, where $Q=p^n$, $q=p^k$ and $p$ is a prime, arises in many different contexts including finite geometry, the inverse Galois problem \cite{ACZ2000}, the construction of difference sets with Singer parameters \cite{DD2004}, determining cross-correlation between $m$-sequences \cite{DOBBERTIN2006} and to construct error correcting codes \cite{Bracken2009}, cryptographic APN functions \cite{BTT2014,Budaghyan-Carlet_2006}, designs \cite{Tang_2019}, as well as to speed up the index calculus method for computing discrete logarithms on finite fields \cite{GGGZ2013,GGGZ2013+} and on algebraic curves \cite{M2014}. Subsequently, in \cite{Bluher2004,HK2008,HK2010,BTT2014,Bluher2016,KM2019,CMPZ2019,MS2019,KCM19}, the $\GF{Q}$-zeros of $P_a(X)$ have been studied. In \cite{Bluher2004}, it was shown that the possible values of the number of the zeros that $P_a(X)$ has in $\GF{Q}$ is $0$, $1$, $2$ or $p^{\gcd(n, k)}+1$. Some criteria for the number of the $\GF{Q}$-zeros of $P_a(x)$ were found in \cite{HK2008,HK2010,BTT2014,KM2019,MS2019}. However, while the ultimate goal is to explicit all the $\GF{Q}$-zeros, even in the case $p=2$, it was solved only under the condition $\gcd(n, k)=1$ \cite{KM2019}. In this article, we discuss this equation without any restriction on $p$ and $\gcd(n,k)$. In \cite{KCM19}, for the cases of one or two $\GF{Q}$-zeros, explicit expressions for these rational zeros in terms of $a$ were provided, but for the case of $p^{\gcd(n, k)}+1$ $\GF{Q}-$ zeros it was remained open to explicitly compute the zeros. This paper solves the remained problem, thus now the equation $X^{p^k+1}+X+a=0$ over $\GF{p^n}$ is completely solved for any prime $p$, any integers $n$ and $k$. △ Less

Submitted 4 January, 2021; originally announced January 2021.

Comments: arXiv admin note: text overlap with arXiv:1912.12648

MSC Class: 12E05; 12E12; 12E10

Preimages of $p-$Linearized Polynomials over $\GF{p}$

Authors: Kwang Ho Kim, Sihem Mesnager, Jong Hyok Choe, Dok Nam Lee

Abstract: Linearized polynomials over finite fields have been intensively studied over the last several decades. Interesting new applications of linearized polynomials to coding theory and finite geometry have been also highlighted in recent years. Let $p$ be any prime. Recently, preimages of the $p-$linearized polynomials $\sum_{i=0}^{\frac kl-1} X^{p^{li}}$ and… ▽ More Linearized polynomials over finite fields have been intensively studied over the last several decades. Interesting new applications of linearized polynomials to coding theory and finite geometry have been also highlighted in recent years. Let $p$ be any prime. Recently, preimages of the $p-$linearized polynomials $\sum_{i=0}^{\frac kl-1} X^{p^{li}}$ and $\sum_{i=0}^{\frac kl-1} (-1)^i X^{p^{li}}$ were explicitly computed over $\GF{p^n}$ for any $n$. This paper extends that study to $p-$linearized polynomials over $\GF{p}$, i.e., polynomials of the shape $$L(X)=\sum_{i=0}^t α_i X^{p^i}, α_i\in\GF{p}.$$ Given a $k$ such that $L(X)$ divides $X-X^{p^k}$, the preimages of $L(X)$ can be explicitly computed over $\GF{p^n}$ for any $n$. △ Less

Submitted 22 November, 2020; originally announced November 2020.

MSC Class: 11D04; 12E05; 12E12

Highly-scalable stochastic neuron based on Ovonic Threshold Switch (OTS) and its applications in Restricted Boltzmann Machine (RBM)

Authors: Seong-il Im, Hye** Lee, Jaesang Lee, Jae-Seung Jeong, Joon Young Kwak, Keunsu Kim, Jeong Ho Cho, Hyunsu Ju, Suyoun Lee

Abstract: Interest in Restricted Boltzmann Machine (RBM) is growing as a generative stochastic artificial neural network to implement a novel energy-efficient machine-learning (ML) technique. For a hardware implementation of the RBM, an essential building block is a reliable stochastic binary neuron device that generates random spikes following the Boltzmann distribution. Here, we propose a highly-scalable… ▽ More Interest in Restricted Boltzmann Machine (RBM) is growing as a generative stochastic artificial neural network to implement a novel energy-efficient machine-learning (ML) technique. For a hardware implementation of the RBM, an essential building block is a reliable stochastic binary neuron device that generates random spikes following the Boltzmann distribution. Here, we propose a highly-scalable stochastic neuron device based on Ovonic Threshold Switch (OTS) which utilizes the random emission and capture process of traps as the source of stochasticity. The switching probability is well described by the Boltzmann distribution, which can be controlled by operating parameters. As a candidate for a true random number generator (TRNG), it passes 15 among the 16 tests of the National Institute of Standards and Technology (NIST) Statistical Test Suite (Special Publication 800-22). In addition, the recognition task of handwritten digits (MNIST) is demonstrated using a simulated RBM network consisting of the proposed device with a maximum recognition accuracy of 86.07 %. Furthermore, reconstruction of images is successfully demonstrated using images contaminated with noises, resulting in images with the noise removed. These results show the promising properties of OTS-based stochastic neuron devices for applications in RBM systems. △ Less

Submitted 21 October, 2020; originally announced October 2020.

Ancient solutions to the Ricci flow with isotropic curvature conditions

Authors: Jae Ho Cho, Yu Li

Abstract: We show that every $n$-dimensional, $κ$-noncollapsed, noncompact, complete ancient solution to the Ricci flow with uniformly PIC for $n=4$ or $n\ge 12$ has weakly PIC$_2$ and bounded curvature. Combining this with earlier results, we prove that any such solution is isometric to either a family of shrinking cylinders (or a quotient thereof) or the Bryant soliton. Also, we classify all complex 2-dim… ▽ More We show that every $n$-dimensional, $κ$-noncollapsed, noncompact, complete ancient solution to the Ricci flow with uniformly PIC for $n=4$ or $n\ge 12$ has weakly PIC$_2$ and bounded curvature. Combining this with earlier results, we prove that any such solution is isometric to either a family of shrinking cylinders (or a quotient thereof) or the Bryant soliton. Also, we classify all complex 2-dimensional, $κ$-noncollapsed, complete ancient solutions to the Kähler Ricci flow with weakly PIC. △ Less

Submitted 24 May, 2020; originally announced May 2020.

Solving Some Affine Equations over Finite Fields

Authors: Sihem Mesnager, Kwang Ho Kim, Jong Hyok Choe, Dok Nam Lee

Abstract: Let $l$ and $k$ be two integers such that $l|k$. Define $T_l^k(X):=X+X^{p^l}+\cdots+X^{p^{l(k/l-2)}}+X^{p^{l(k/l-1)}}$ and $S_l^k(X):=X-X^{p^l}+\cdots+(-1)^{(k/l-1)}X^{p^{l(k/l-1)}}$, where $p$ is any prime. This paper gives explicit representations of all solutions in $\GF{p^n}$ to the affine equations $T_l^{k}(X)=a$ and $S_l^{k}(X)=a$, $a\in \GF{p^n}$. For the case $p=2$ that was solved very r… ▽ More Let $l$ and $k$ be two integers such that $l|k$. Define $T_l^k(X):=X+X^{p^l}+\cdots+X^{p^{l(k/l-2)}}+X^{p^{l(k/l-1)}}$ and $S_l^k(X):=X-X^{p^l}+\cdots+(-1)^{(k/l-1)}X^{p^{l(k/l-1)}}$, where $p$ is any prime. This paper gives explicit representations of all solutions in $\GF{p^n}$ to the affine equations $T_l^{k}(X)=a$ and $S_l^{k}(X)=a$, $a\in \GF{p^n}$. For the case $p=2$ that was solved very recently in \cite{MKCL2019}, the result of this paper reveals another solution. △ Less

Submitted 12 February, 2020; originally announced February 2020.

Infrared study of carrier scattering mechanism in ion-gated graphene

Authors: Kwangnam Yu, Jiwon Jeon, Jiho Kim, Chang Won Oh, Yongseok Yoon, Beom Joon Kim, Jeong Ho Cho, E. J. Choi

Abstract: We performed infrared transmission experiment on ion-gel gated graphene and measured carrier scattering rate g as function of carrier density n over wide range up to n=2E13 cm-2. The g exhibits a rapid decreases along with the gating followed by persistent increases on further carrier do**. This behavior of g(n) demonstrates that carrier is scattered dominantly by the two scattering mechanisms,… ▽ More We performed infrared transmission experiment on ion-gel gated graphene and measured carrier scattering rate g as function of carrier density n over wide range up to n=2E13 cm-2. The g exhibits a rapid decreases along with the gating followed by persistent increases on further carrier do**. This behavior of g(n) demonstrates that carrier is scattered dominantly by the two scattering mechanisms, namely, charged impurity (CI) scattering and short-range disorder (SR) scattering, with additional minor scattering from substrate phonon (SPP). We can determine the absolute strengths of all the scattering channels by fitting the g(n) data and unveils the complete n-dependent map of the scattering mechanisms g(n)=gCI(n)+gSR(n)+gSPP(n). The gCI(n) and gSR(n) are larger than those of SiO2$-gated graphene by 1.8 times, which elucidates the dual role of the ion-gel layer as a CI-scatterer and simultaneously a SR-scatterer to graphene. Additionally we show that freezing of IG at low-T (~200 K) does not cause any change to the carrier scattering. △ Less

Submitted 7 November, 2019; originally announced November 2019.

Comments: 5 pages, 4 figures

Journal ref: Appl. Phys. Lett. 114, 083503 (2019)

Emergence of Multiscale Dynamics in Colloidal Gels

Authors: Jae Hyung Cho, Roberto Cerbino, Irmgard Bischofberger

Abstract: To gain insight into the kinetics of colloidal gel evolution at low particle volume fractions $φ$, we utilize differential dynamic microscopy to investigate particle aggregation, geometric percolation, and the subsequent transition to nonergodic dynamics. We report the emergence of unexpectedly rich multiscale dynamics upon the onset of nonergodicity, which separates the wave vectors $q$ into thre… ▽ More To gain insight into the kinetics of colloidal gel evolution at low particle volume fractions $φ$, we utilize differential dynamic microscopy to investigate particle aggregation, geometric percolation, and the subsequent transition to nonergodic dynamics. We report the emergence of unexpectedly rich multiscale dynamics upon the onset of nonergodicity, which separates the wave vectors $q$ into three different regimes. In the high-$q$ domain, the gel exhibits $φ$-independent internal vibrations of fractal clusters. The intermediate-$q$ domain is dominated by density fluctuations at the length scale of the clusters, as evidenced by the $q$-independence of the relaxation time $τ$. In the low-$q$ domain, the scaling of $τ$ as $q^{-3}$ suggests that the network appears homogeneous. The transitions between these three regimes introduce two characteristic length scales, distinct from the cluster size. △ Less

Submitted 25 February, 2020; v1 submitted 4 November, 2019; originally announced November 2019.

Journal ref: Phys. Rev. Lett. 124, 088005 (2020)

On the Efficacy of Knowledge Distillation

Authors: Jang Hyun Cho, Bharath Hariharan

Abstract: In this paper, we present a thorough evaluation of the efficacy of knowledge distillation and its dependence on student and teacher architectures. Starting with the observation that more accurate teachers often don't make good teachers, we attempt to tease apart the factors that affect knowledge distillation performance. We find crucially that larger models do not often make better teachers. We sh… ▽ More In this paper, we present a thorough evaluation of the efficacy of knowledge distillation and its dependence on student and teacher architectures. Starting with the observation that more accurate teachers often don't make good teachers, we attempt to tease apart the factors that affect knowledge distillation performance. We find crucially that larger models do not often make better teachers. We show that this is a consequence of mismatched capacity, and that small students are unable to mimic large teachers. We find typical ways of circumventing this (such as performing a sequence of knowledge distillation steps) to be ineffective. Finally, we show that this effect can be mitigated by stop** the teacher's training early. Our results generalize across datasets and models. △ Less

Submitted 3 October, 2019; originally announced October 2019.

Comments: 13 pages, including Appendix

Journal ref: ICCV 2019

Solutions of $x^{q^k}+\cdots+x^{q}+x=a$ in $GF{2^n}$

Authors: Kwang Ho Kim, Jong Hyok Choe, Dok Nam Lee, Dae Song Go, Sihem Mesnager

Abstract: Though it is well known that the roots of any affine polynomial over a finite field can be computed by a system of linear equations by using a normal base of the field, such solving approach appears to be difficult to apply when the field is fairly large. Thus, it may be of great interest to find an explicit representation of the solutions independently of the field base. This was previously done… ▽ More Though it is well known that the roots of any affine polynomial over a finite field can be computed by a system of linear equations by using a normal base of the field, such solving approach appears to be difficult to apply when the field is fairly large. Thus, it may be of great interest to find an explicit representation of the solutions independently of the field base. This was previously done only for quadratic equations over a binary finite field. This paper gives an explicit representation of solutions for a much wider class of affine polynomials over a binary prime field. △ Less

Submitted 25 May, 2019; originally announced May 2019.

Probing Out-of-Plane Charge Transport in Black Phosphorus with Graphene-Contacted Vertical Field-Effect Transistors

Authors: Junmo Kang, Deep Jariwala, Christopher R. Ryder, Spencer A. Wells, Yongsuk Choi, Euyheon Hwang, Jeong Ho Cho, Tobin J. Marks, Mark C. Hersam

Abstract: Black phosphorus (BP) has recently emerged as a promising narrow band gap layered semiconductor with optoelectronic properties that bridge the gap between semi-metallic graphene and wide band gap transition metal dichalcogenides such as MoS2. To date, BP field-effect transistors have utilized a lateral geometry with in-plane transport dominating device characteristics. In contrast, we present here… ▽ More Black phosphorus (BP) has recently emerged as a promising narrow band gap layered semiconductor with optoelectronic properties that bridge the gap between semi-metallic graphene and wide band gap transition metal dichalcogenides such as MoS2. To date, BP field-effect transistors have utilized a lateral geometry with in-plane transport dominating device characteristics. In contrast, we present here a vertical field-effect transistor geometry based on a graphene/BP van der Waals heterostructure. The resulting device characteristics include high on-state current densities (> 1600 A/cm2) and current on/off ratios exceeding 800 at low temperature. Two distinct charge transport mechanisms are identified, which are dominant for different regimes of temperature and gate voltage. In particular, the Schottky barrier between graphene and BP determines charge transport at high temperatures and positive gate voltages, whereas tunneling dominates at low temperatures and negative gate voltages. These results elucidate out-of-plane electronic transport in BP, and thus have implications for the design and operation of BP-based van der Waals heterostructures. △ Less

Submitted 13 March, 2016; originally announced March 2016.

Comments: 4 figures and supporting information, Published in Nano Letters (2016)

Apparent Positions of Planets

Authors: Chol Jun Kim, Kyong Il Pak, Sin Chol Hwang, Chol Min Choe, ** Hyok Choe, Ui Ri Mun

Abstract: The apparent positions of planets are determined by means of the fundamental ephemerides, the precession-nutation models of the Earth, the gravitational effects and aberrations et al. Around 2000, many astrometrical conceptions, models and theories had been newly defined and updated:for the fiducial celestial reference system, the ICRS is introduced, the fundamental ephemerides - DE405/LE405 et al… ▽ More The apparent positions of planets are determined by means of the fundamental ephemerides, the precession-nutation models of the Earth, the gravitational effects and aberrations et al. Around 2000, many astrometrical conceptions, models and theories had been newly defined and updated:for the fiducial celestial reference system, the ICRS is introduced, the fundamental ephemerides - DE405/LE405 et al.,precession-nutation model - IAU 2000A/IAU 2006 model. Using the traditional algorithm and the updated models, we develop the system of calculating the apparent positions of planets. The results are compared with the Astronomical Almanac and proved in their correctness. △ Less

Submitted 23 August, 2015; originally announced August 2015.

Comments: 9 pages, 3 figures

Joint Transmitter and Receiver Optimization for Improper-Complex Second-Order Stationary Data Sequence

Authors: Jeongho Yeo, Joon Ho Cho, James S. Lehnert

Abstract: In this paper, the transmission of an improper-complex second-order stationary data sequence is considered over a strictly band-limited frequency-selective channel. It is assumed that the transmitter employs linear modulation and that the channel output is corrupted by additive proper-complex cyclostationary noise. Under the average transmit power constraint, the problem of minimizing the mean-squ… ▽ More In this paper, the transmission of an improper-complex second-order stationary data sequence is considered over a strictly band-limited frequency-selective channel. It is assumed that the transmitter employs linear modulation and that the channel output is corrupted by additive proper-complex cyclostationary noise. Under the average transmit power constraint, the problem of minimizing the mean-squared error at the output of a widely linear receiver is formulated in the time domain to find the optimal transmit and receive waveforms. The optimization problem is converted into a frequency-domain problem by using the vectorized Fourier transform technique and put into the form of a double minimization. First, the widely linear receiver is optimized that requires, unlike the linear receiver design with only one waveform, the design of two receive waveforms. Then, the optimal transmit waveform for the linear modulator is derived by introducing the notion of the impropriety frequency function of a discrete-time random process and by performing a line search combined with an iterative algorithm. The optimal solution shows that both the periodic spectral correlation due to the cyclostationarity and the symmetric spectral correlation about the origin due to the impropriety are well exploited. △ Less

Submitted 25 May, 2014; originally announced May 2014.

Asymptotic FRESH Properizer for Block Processing of Improper-Complex Second-Order Cyclostationary Random Processes

Authors: Jeongho Yeo, Joon Ho Cho

Abstract: In this paper, the block processing of a discrete-time (DT) improper-complex second-order cyclostationary (SOCS) random process is considered. In particular, it is of interest to find a pre-processing operation that enables computationally efficient near-optimal post-processing. An invertible linear-conjugate linear (LCL) operator named the DT FREquency Shift (FRESH) properizer is first proposed.… ▽ More In this paper, the block processing of a discrete-time (DT) improper-complex second-order cyclostationary (SOCS) random process is considered. In particular, it is of interest to find a pre-processing operation that enables computationally efficient near-optimal post-processing. An invertible linear-conjugate linear (LCL) operator named the DT FREquency Shift (FRESH) properizer is first proposed. It is shown that the DT FRESH properizer converts a DT improper-complex SOCS random process input to an equivalent DT proper-complex SOCS random process output by utilizing the information only about the cycle period of the input. An invertible LCL block processing operator named the asymptotic FRESH properizer is then proposed that mimics the operation of the DT FRESH properizer but processes a finite number of consecutive samples of a DT improper-complex SOCS random process. It is shown that the output of the asymptotic FRESH properizer is not proper but asymptotically proper and that its frequency-domain covariance matrix converges to a highly-structured block matrix with diagonal blocks as the block size tends to infinity. Two representative estimation and detection problems are presented to demonstrate that asymptotically optimal low-complexity post-processors can be easily designed by exploiting these asymptotic second-order properties of the output of the asymptotic FRESH properizer. △ Less

Submitted 27 April, 2013; originally announced April 2013.

Comments: 42 pages, 13 figures

Sum-Rate Optimal Multi-Code CDMA Systems: An Equivalence Result

Authors: Yeo Hun Yun, Joon Ho Cho

Abstract: In this paper, the sum rate of a multi-code CDMA system with asymmetric-power users is maximized, given a processing gain and a power profile of users. Unlike the sum-rate maximization for a single-code CDMA system, the optimization requires the joint optimal distribution of each user's power to its multiple data streams as well as the optimal design of signature sequences. The crucial step is to… ▽ More In this paper, the sum rate of a multi-code CDMA system with asymmetric-power users is maximized, given a processing gain and a power profile of users. Unlike the sum-rate maximization for a single-code CDMA system, the optimization requires the joint optimal distribution of each user's power to its multiple data streams as well as the optimal design of signature sequences. The crucial step is to establish an equivalence of the multi-code CDMA system to restricted FDMA and TDMA systems. The CDMA system has upper limits on the numbers of multi-codes of users, while the FDMA and the TDMA systems have upper limits on the bandwidths and the duty cycles of users, respectively, in addition to total bandwidth constraint. The equivalence facilitates the complete characterization of the maximum sum rate of the multi-code CDMA system and also provides new insights into the single- and the multi-code CDMA systems in terms of the parameters of the equivalent FDMA and TDMA systems. △ Less

Submitted 27 November, 2012; originally announced November 2012.

Comments: 66 pages, 7 figures

Infrared Hall conductivity of Na$_{0.7}$CoO$_2$

Authors: E. J. Choi, S. H. Jung, J. H. Noh, A. Zimmers, D. Schmadel, H. D. Drew, J. Y. Son, J. H. Cho

Abstract: We report infrared Hall conductivity $σ_{xy}(ω)$ of Na$_{0.7}$CoO$_2$ thin films determined from Faraday rotation angle $θ_{F}$ measurements. $σ_{xy}(ω)$ exhibits two types of hole conduction, Drude and incoherent carriers. The coherent Drude carrier shows a large renormalized mass and Fermi liquid-like behavior of Hall scattering rate, $γ_{H} \sim aT^{2}$. The spectral weight is suppressed and… ▽ More We report infrared Hall conductivity $σ_{xy}(ω)$ of Na$_{0.7}$CoO$_2$ thin films determined from Faraday rotation angle $θ_{F}$ measurements. $σ_{xy}(ω)$ exhibits two types of hole conduction, Drude and incoherent carriers. The coherent Drude carrier shows a large renormalized mass and Fermi liquid-like behavior of Hall scattering rate, $γ_{H} \sim aT^{2}$. The spectral weight is suppressed and disappears at T = 120K. The incoherent carrier response is centered at mid-IR frequency and shifts to lower energy with increasing T. Infrared Hall constant is positive and almost independent of temperature in sharp contrast with the dc-Hall constant. △ Less

Submitted 3 November, 2006; v1 submitted 1 November, 2006; originally announced November 2006.

Comments: 5 Pages, 5 Figures. Author list corrected in metadata only, paper is unchanged

Kinetically-controlled thin-film growth of layered $β$- and $γ-$Na$_{x}$CoO$_{2}$ cobaltate

Authors: J. Y. Son, Bog G. Kim, J. H. Cho

Abstract: We report growth characteristics of epitaxial $β$-Na$_{0.6}$CoO$_{2}$ and $γ$-Na$_{0.7}$CoO$_{2}$ thin films on (001) sapphire substrates grown by pulsed-laser deposition. Reduction of deposition rate could change structure of Na$_{x}$CoO$_{2}$ thin film from $β$-phase with island growth mode to $γ$-phase with layer-by-layer growth mode. The $γ$-Na$_{0.7}$CoO$_{2}$ thin film exhibits spiral surf… ▽ More We report growth characteristics of epitaxial $β$-Na$_{0.6}$CoO$_{2}$ and $γ$-Na$_{0.7}$CoO$_{2}$ thin films on (001) sapphire substrates grown by pulsed-laser deposition. Reduction of deposition rate could change structure of Na$_{x}$CoO$_{2}$ thin film from $β$-phase with island growth mode to $γ$-phase with layer-by-layer growth mode. The $γ$-Na$_{0.7}$CoO$_{2}$ thin film exhibits spiral surface growth with multiterraced islands and highly crystallized texture compared to that of the $β$-Na$_{0.6}$CoO$_{2}$ thin film. This heterogeneous epitaxial film growth can give opportunity of strain effect of physical properties and growth dynamics of Na$_{x}$CoO$_{2}$ as well as subtle nature of structural change. △ Less

Submitted 11 May, 2005; originally announced May 2005.

Comments: accepted for publication in Applied Physics Letters

Spin chirality on a two-dimensional frustrated lattice

Authors: D. Grohol, K. Matan, J. H. Cho, S. -H. Lee, J. W. Lynn, D. G. Nocera, Y. S. Lee

Abstract: The collective behavior of interacting magnetic moments can be strongly influenced by the topology of the underlying lattice. In geometrically frustrated spin systems, interesting chiral correlations may develop that are related to the spin arrangement on triangular plaquettes. We report a study of the spin chirality on a two-dimensional geometrically frustrated lattice. Our new chemical synthes… ▽ More The collective behavior of interacting magnetic moments can be strongly influenced by the topology of the underlying lattice. In geometrically frustrated spin systems, interesting chiral correlations may develop that are related to the spin arrangement on triangular plaquettes. We report a study of the spin chirality on a two-dimensional geometrically frustrated lattice. Our new chemical synthesis methods allow us to produce large single crystal samples of KFe3(OH)6(SO4)2, an ideal Kagome lattice antiferromagnet. Combined thermodynamic and neutron scattering measurements reveal that the phase transition to the ordered ground-state is unusual. At low temperatures, application of a magnetic field induces a transition between states with different non-trivial spin-textures. △ Less

Submitted 5 April, 2005; v1 submitted 5 April, 2005; originally announced April 2005.

Comments: 7 pages, 4 figures

Journal ref: Nature Materials Vol. 4 323-328 April 2005

Structural transition in NaxCoO2 with x near 0.75 due to Na rearrangement

Authors: Q. Huang, B. Khaykovich, F. C. Chou, J. H. Cho, J. W. Lynn, Y. S. Lee

Abstract: We report neutron powder diffraction measurements on a series of NaxCoO2 samples with x near 0.75 which were prepared under different synthesis conditions. Two different crystal structures for the samples are observed at room temperature. The two structures differ primarily by a shift of a large fraction of the Na ions from a high-symmetry position to a lower-symmetry position. Close inspection… ▽ More We report neutron powder diffraction measurements on a series of NaxCoO2 samples with x near 0.75 which were prepared under different synthesis conditions. Two different crystal structures for the samples are observed at room temperature. The two structures differ primarily by a shift of a large fraction of the Na ions from a high-symmetry position to a lower-symmetry position. Close inspection of the refinement parameters indicates that the presence of either structure depends sensitively on the Na content x, with x = 0.75 as the critical concentration separating the two phases. By raising the temperature to around T = 323 K, the high-symmetry structure can be converted to the lower-symmetry structure. The transition is reversible, but there is significant hysteresis. We discuss the effects of this structural transition on the bulk magnetic and transport properties. △ Less

Submitted 28 May, 2004; originally announced May 2004.

Comments: 7 pages, 7 figures

Journal ref: Phys. Rev. B 70, 134115 (2004)

Electrochemical De-intercalation, Oxygen Non-stoichiometry, and Crystal Growth of NaxCoO2-d

Authors: F. C. Chou, E. T. Abel, J. H. Cho, Y. S. Lee

Abstract: We report a detailed study of de-intercalation of Na from the compound NaxCoO2-d using an electrochemical technique. We find evidence for stable phases with Na contents near the fractions ~1/3, 1/2, 5/8, 2/3, and 3/4. Details regarding the floating-zone crystal growth of Na0.75CoO2 single crystals are discussed as well as results from magnetic susceptibility measurements. We observe the presence… ▽ More We report a detailed study of de-intercalation of Na from the compound NaxCoO2-d using an electrochemical technique. We find evidence for stable phases with Na contents near the fractions ~1/3, 1/2, 5/8, 2/3, and 3/4. Details regarding the floating-zone crystal growth of Na0.75CoO2 single crystals are discussed as well as results from magnetic susceptibility measurements. We observe the presence of significant oxygen deficiencies in powder samples of Na0.75CoO2-d prepared in air, but not in single crystal samples prepared in an oxygen atmosphere. The oxygen deficiencies in a Na0.75CoO2-d sample with d ~ 0.08 remain even after electrochemically de-intercalating to Na0.3CoO2-d. △ Less

Submitted 7 May, 2004; originally announced May 2004.

Comments: 6 pages, 5 figures

Magnetic susceptibility study of hydrated and non-hydrated NaxCoO2-yH2O single crystals

Authors: F. C. Chou, J. H. Cho, Y. S. Lee

Abstract: We have measured the magnetic susceptibility of single crystal samples of non-hydrated NaxCoO2 (x ~ 0.75, 0.67, 0.5, and 0.3) and hydrated Na0.3CoO2-yH2O (y ~ 0, 0.6, 1.3). Our measurements reveal considerable anisotropy between the susceptibilities with H||c and H||ab. The derived anisotropic g-factor ratio (g_ab/g_c) decreases significantly as the composition is changed from the Curie-Weiss me… ▽ More We have measured the magnetic susceptibility of single crystal samples of non-hydrated NaxCoO2 (x ~ 0.75, 0.67, 0.5, and 0.3) and hydrated Na0.3CoO2-yH2O (y ~ 0, 0.6, 1.3). Our measurements reveal considerable anisotropy between the susceptibilities with H||c and H||ab. The derived anisotropic g-factor ratio (g_ab/g_c) decreases significantly as the composition is changed from the Curie-Weiss metal with x = 0.75 to the paramagnetic metal with x = 0.3. Fully hydrated Na0.3CoO2-1.3H2O samples have a larger susceptibility than non-hydrated Na0.3CoO2 samples, as well as a higher degree of anisotropy. In addition, the fully hydrated compound contains a small additional fraction of anisotropic localized spins. △ Less

Submitted 1 June, 2004; v1 submitted 2 April, 2004; originally announced April 2004.

Comments: 6 pages, 5 figures

Thermodynamic and Transport Measurements on Superconducting Na_xCoO_2 + yH_2O Single Crystals Prepared by Electrochemical De-intercalation

Authors: F. C. Chou, J. H. Cho, Patrick A. Lee, E. T. Abel, K. Matan, Young S. Lee

Abstract: Superconducting single crystal samples of Na_xCoO_2 + yH_2O have been produced using an electrochemical technique which dispenses with the usual bromine chemical de-intercalation step and permits much more precise control of the Na content. After careful hydration, nearly single-phase crystals have been obtained in which over 90% of the sample's volume corresponds to the superconducting Na_{0.3}… ▽ More Superconducting single crystal samples of Na_xCoO_2 + yH_2O have been produced using an electrochemical technique which dispenses with the usual bromine chemical de-intercalation step and permits much more precise control of the Na content. After careful hydration, nearly single-phase crystals have been obtained in which over 90% of the sample's volume corresponds to the superconducting Na_{0.3}CoO_2 + 1.3H_2O structure. Susceptibility and specific heat measurements confirm that bulk superconductivity has been achieved. The extracted normal state density of states indicates Fermi-liquid behavior with strong mass enhancement and a modest Wilson ratio. Measurements of H_{c2} for H||c and H||ab reveal significant anisotropy. The estimated value of H_{c2} for H||c yields a coherence length of ~100 A, consistent with an extremely narrow bandwidth. △ Less

Submitted 17 October, 2003; v1 submitted 26 June, 2003; originally announced June 2003.

Comments: 5 pages; 5 figures; new data in figures; revised and updated text

Spin dynamics in hole-doped two-dimensional S=1/2 Heisenberg antiferromagnets: ^{63}Cu NQR relaxation in La_{2-x}Sr_xCuO_4 for $x\leq 0.04$

Authors: P. Carretta, F. Tedoldi, A. Rigamonti, F. Galli, F. Borsa, J. H. Cho, D. C. Johnston

Abstract: The effects on the correlated Cu^{2+} S = 1/2 spin dynamics in the paramagnetic phase of La_{2-x}Sr_xCuO_4 (for $x \lesssim 0.04$) due to the injection of holes are studied by means of ^{63}Cu NQR spin-lattice relaxation time T_1 measurements. The results are discussed in the framework of the connection between T_1 and the in-plane magnetic correlation length $ξ_{2D}(x,T)$. It is found that at h… ▽ More The effects on the correlated Cu^{2+} S = 1/2 spin dynamics in the paramagnetic phase of La_{2-x}Sr_xCuO_4 (for $x \lesssim 0.04$) due to the injection of holes are studied by means of ^{63}Cu NQR spin-lattice relaxation time T_1 measurements. The results are discussed in the framework of the connection between T_1 and the in-plane magnetic correlation length $ξ_{2D}(x,T)$. It is found that at high temperatures the system remains in the renormalized classical regime, with a spin stiffness constant $ρ_s(x)$ reduced by small do** to an extent larger than the one due to Zn do**. For $x\gtrsim 0.02$ the effect of do** on $ρ_s(x)$ appears to level off. The values for $ρ_s(x)$ derived from T_1 for $T\gtrsim 500$ K are much larger than the ones estimated from the temperature behavior of sublattice magnetization in the ordered phase ($T\leq T_N$). It is argued that these features are consistent with the hypothesis of formation of stripes of microsegregated holes. △ Less

Submitted 31 March, 1999; originally announced March 1999.

Comments: 10 pages, 3 figures

Journal ref: Eur. Phys. J. B 10, 233-236 (1999)

Spectral evolution in (Ca,Sr)RuO_3 near the Mott-Hubbard transition

Authors: J. S. Ahn, J. Bak, H. S. Choi, T. W. Noh, J. E. Han, Yunkyu Bang, J. H. Cho, Q. X. Xia

Abstract: We investigated the optical properties of (Ca,Sr)RuO_3 films on the borderline of a metal-insulator (M-I) transition. Our results show all of the predicted characteristics for a metallic Mott-Hubbard system, including (i) a mass enhancement in dc-limit, (ii) an U/2 excitation, and (iii) an U excitation. Also, a self-consistency is found within the Gutzwiller-Brinkman-Rice picture for the Mott tr… ▽ More We investigated the optical properties of (Ca,Sr)RuO_3 films on the borderline of a metal-insulator (M-I) transition. Our results show all of the predicted characteristics for a metallic Mott-Hubbard system, including (i) a mass enhancement in dc-limit, (ii) an U/2 excitation, and (iii) an U excitation. Also, a self-consistency is found within the Gutzwiller-Brinkman-Rice picture for the Mott transition. Our finding displays that electron correlation should be important even in 4d materials. △ Less

Submitted 14 July, 1998; originally announced July 1998.

Comments: REVTEX 4 pages, 5 EPS figures, submitted to Phys. Rev. Lett