-
Quasiperiodic [110] Symmetric Tilt FCC Grain Boundaries
Authors:
Wenwen Zou,
Juan Zhang,
Jie Xu,
Kai Jiang
Abstract:
In this work, we investigate [110] symmetric tilt FCC grain boundaries (GBs) by a recently developed approach for quasiperiodic interfaces using the Landau-Brazovskii model. On special tilt angles associated with quadratic algebraic numbers, quasiperiodic GBs exhibit generalized Fibonacci substitution rules. The transition mechanism from quasiperiodic GBs to periodic GBs is explored through sphere…
▽ More
In this work, we investigate [110] symmetric tilt FCC grain boundaries (GBs) by a recently developed approach for quasiperiodic interfaces using the Landau-Brazovskii model. On special tilt angles associated with quadratic algebraic numbers, quasiperiodic GBs exhibit generalized Fibonacci substitution rules. The transition mechanism from quasiperiodic GBs to periodic GBs is explored through sphere offsets and spectral coalescence. We also propose an accurate method to calculate the GB energy for arbitrary tilt angle and analyze the factors affecting the GB energy. The revealing GB energy change is continue along with tilt angle except periodic GBs.
△ Less
Submitted 5 June, 2024;
originally announced June 2024.
-
Atomic-scale tunable phonon transport at tailored grain boundaries
Authors:
Xiaowang Wang,
Chaitanya A. Gadre,
Runqing Yang,
Wanjuan Zou,
Xing Bin,
Christopher Addiego,
Toshihiro Aoki,
Yujie Quan,
Wei-Tao Peng,
Yifeng Huang,
Chaojie Du,
Mingjie Xu,
Xingxu Yan,
Ruqian Wu,
Shyue ** Ong,
Bolin Liao,
Penghui Cao,
Xiaoqing Pan
Abstract:
Manipulating thermal properties in materials has been of fundamental importance for advancing innovative technologies. Heat carriers such as phonons are impeded by breaking crystal symmetry or periodicity. Notable methods of impeding the phonon propagation include varying the density of defects, interfaces, and nanostructures, as well as changing composition. However, a robust link between the ind…
▽ More
Manipulating thermal properties in materials has been of fundamental importance for advancing innovative technologies. Heat carriers such as phonons are impeded by breaking crystal symmetry or periodicity. Notable methods of impeding the phonon propagation include varying the density of defects, interfaces, and nanostructures, as well as changing composition. However, a robust link between the individual nanoscale defect structures, phonon states, and macroscopic thermal conductivity is lacking. Here we reveal from nanoscale structure-phonon mechanisms on how the grain boundary (GB) tilt and twist angles fundamentally drive the changes in atom rearrangements, exotic vibrational states, and finally macroscopic heat transport at different bicrystal strontium titanate GBs using emerging atomic resolution vibrational spectroscopy. The 10 deg and 22 deg tilt GBs exhibit reduced phonon populations by 54% and 16% compared to the bulk value, respectively, consistent with measured thermal conductivities. A tiny twist angle further introduces a fine and local tunning of thermal conductivity by introducing twist induced defects periodically embedded with the tilt induced GB defects. Our results demonstrate that varying the tilt angle coarsely modifies the phonon population along entire GB while varying the twist angle incurs a finer adjustment at periodic locations on the GB. Our study offers a systematic approach to understanding and manipulating cross GB thermal transport of arbitrary GBs predictably and precisely.
△ Less
Submitted 13 May, 2024;
originally announced May 2024.
-
Introduction to dynamical mean-field theory of randomly connected neural networks with bidirectionally correlated couplings
Authors:
Wenxuan Zou,
Hai** Huang
Abstract:
Dynamical mean-field theory is a powerful physics tool used to analyze the typical behavior of neural networks, where neurons can be recurrently connected, or multiple layers of neurons can be stacked. However, it is not easy for beginners to access the essence of this tool and the underlying physics. Here, we give a pedagogical introduction of this method in a particular example of random neural…
▽ More
Dynamical mean-field theory is a powerful physics tool used to analyze the typical behavior of neural networks, where neurons can be recurrently connected, or multiple layers of neurons can be stacked. However, it is not easy for beginners to access the essence of this tool and the underlying physics. Here, we give a pedagogical introduction of this method in a particular example of random neural networks, where neurons are randomly and fully connected by correlated synapses and therefore the network exhibits rich emergent collective dynamics. We also review related past and recent important works applying this tool. In addition, a physically transparent and alternative method, namely the dynamical cavity method, is also introduced to derive exactly the same results. The numerical implementation of solving the integro-differential mean-field equations is also detailed, with an illustration of exploring the fluctuation dissipation theorem.
△ Less
Submitted 7 October, 2023; v1 submitted 15 May, 2023;
originally announced May 2023.
-
Statistical mechanics of continual learning: variational principle and mean-field potential
Authors:
Chan Li,
Zhenye Huang,
Wenxuan Zou,
Hai** Huang
Abstract:
An obstacle to artificial general intelligence is set by continual learning of multiple tasks of different nature. Recently, various heuristic tricks, both from machine learning and from neuroscience angles, were proposed, but they lack a unified theory ground. Here, we focus on continual learning in single-layered and multi-layered neural networks of binary weights. A variational Bayesian learnin…
▽ More
An obstacle to artificial general intelligence is set by continual learning of multiple tasks of different nature. Recently, various heuristic tricks, both from machine learning and from neuroscience angles, were proposed, but they lack a unified theory ground. Here, we focus on continual learning in single-layered and multi-layered neural networks of binary weights. A variational Bayesian learning setting is thus proposed, where the neural networks are trained in a field-space, rather than gradient-ill-defined discrete-weight space, and furthermore, weight uncertainty is naturally incorporated, and modulates synaptic resources among tasks. From a physics perspective, we translate the variational continual learning into Franz-Parisi thermodynamic potential framework, where previous task knowledge acts as a prior and a reference as well. We thus interpret the continual learning of the binary perceptron in a teacher-student setting as a Franz-Parisi potential computation. The learning performance can then be analytically studied with mean-field order parameters, whose predictions coincide with numerical experiments using stochastic gradient descent methods. Based on the variational principle and Gaussian field approximation of internal preactivations in hidden layers, we also derive the learning algorithm considering weight uncertainty, which solves the continual learning with binary weights using multi-layered neural networks, and performs better than the currently available metaplasticity algorithm. Our proposed principled frameworks also connect to elastic weight consolidation, weight-uncertainty modulated learning, and neuroscience inspired metaplasticity, providing a theory-grounded method for the real-world multi-task learning with deep networks.
△ Less
Submitted 20 June, 2023; v1 submitted 6 December, 2022;
originally announced December 2022.
-
Enormous Berry-Curvature-Driven Anomalous Hall Effect in Topological Insulator (Bi,Sb)2Te3 on Ferrimagnetic Europium Iron Garnet beyond 400 K
Authors:
Wei-Jhih Zou,
Meng-Xin Guo,
Jyun-Fong Wong,
Zih-** Huang,
Jui-Min Chia,
Wei-Nien Chen,
Sheng-Xin Wang,
Keng-Yung Lin,
Lawrence Boyu Young,
Yen-Hsun Glen Lin,
Mohammad Yahyavi,
Chien-Ting Wu,
Horng-Tay Jeng,
Shang-Fan Lee,
Tay-Rong Chang,
Minghwei Hong,
Jueinai Kwo
Abstract:
To realize the quantum anomalous Hall effect (QAHE) at elevated temperatures, the approach of magnetic proximity effect (MPE) was adopted to break the time-reversal symmetry in the topological insulator (Bi0.3Sb0.7)2Te3 (BST) based heterostructures with a ferrimagnetic insulator europium iron garnet (EuIG) of perpendicular magnetic anisotropy. Here we demonstrate phenomenally large anomalous Hall…
▽ More
To realize the quantum anomalous Hall effect (QAHE) at elevated temperatures, the approach of magnetic proximity effect (MPE) was adopted to break the time-reversal symmetry in the topological insulator (Bi0.3Sb0.7)2Te3 (BST) based heterostructures with a ferrimagnetic insulator europium iron garnet (EuIG) of perpendicular magnetic anisotropy. Here we demonstrate phenomenally large anomalous Hall resistance (RAHE) exceeding 8 Ω (\r{ho}AHE of 3.2 μΩ*cm) at 300 K and sustaining to 400 K in 35 BST/EuIG samples, surpassing the past record of 0.28 Ω (\r{ho}AHE of 0.14 μΩ*cm) at 300 K. The remarkably large RAHE as attributed to an atomically abrupt, Fe-rich interface between BST and EuIG. Importantly, the gate dependence of the AHE loops shows no sign change with varying chemical potential. This observation is supported by our first-principles calculations via applying a gradient Zeeman field plus a contact potential on BST. Our calculations further demonstrate that the AHE in this heterostructure is attributed to the intrinsic Berry curvature. Furthermore, for gate-biased 4 nm BST on EuIG, a pronounced topological Hall effect (THE) coexisting with AHE is observed at the negative top-gate voltage up to 15 K. Interface tuning with theoretical calculations has opened up new opportunities to realize topologically distinct phenomena in tailored magnetic TI-based heterostructures.
△ Less
Submitted 30 September, 2021; v1 submitted 30 March, 2021;
originally announced March 2021.
-
Ensemble perspective for understanding temporal credit assignment
Authors:
Wenxuan Zou,
Chan Li,
Hai** Huang
Abstract:
Recurrent neural networks are widely used for modeling spatio-temporal sequences in both nature language processing and neural population dynamics. However, understanding the temporal credit assignment is hard. Here, we propose that each individual connection in the recurrent computation is modeled by a spike and slab distribution, rather than a precise weight value. We then derive the mean-field…
▽ More
Recurrent neural networks are widely used for modeling spatio-temporal sequences in both nature language processing and neural population dynamics. However, understanding the temporal credit assignment is hard. Here, we propose that each individual connection in the recurrent computation is modeled by a spike and slab distribution, rather than a precise weight value. We then derive the mean-field algorithm to train the network at the ensemble level. The method is then applied to classify handwritten digits when pixels are read in sequence, and to the multisensory integration task that is a fundamental cognitive function of animals. Our model reveals important connections that determine the overall performance of the network. The model also shows how spatio-temporal information is processed through the hyperparameters of the distribution, and moreover reveals distinct types of emergent neural selectivity. To provide a mechanistic analysis of the ensemble learning, we first derive an analytic solution of the learning at the infinitely-large-network limit. We then carry out a low-dimensional projection of both neural and synaptic dynamics, analyze symmetry breaking in the parameter space, and finally demonstrate the role of stochastic plasticity in the recurrent computation. Therefore, our study sheds light on mechanisms of how weight uncertainty impacts the temporal credit assignment in recurrent neural networks from the ensemble perspective.
△ Less
Submitted 7 March, 2022; v1 submitted 7 February, 2021;
originally announced February 2021.
-
Data-driven effective model shows a liquid-like deep learning
Authors:
Wenxuan Zou,
Hai** Huang
Abstract:
The geometric structure of an optimization landscape is argued to be fundamentally important to support the success of deep neural network learning. A direct computation of the landscape beyond two layers is hard. Therefore, to capture the global view of the landscape, an interpretable model of the network-parameter (or weight) space must be established. However, the model is lacking so far. Furth…
▽ More
The geometric structure of an optimization landscape is argued to be fundamentally important to support the success of deep neural network learning. A direct computation of the landscape beyond two layers is hard. Therefore, to capture the global view of the landscape, an interpretable model of the network-parameter (or weight) space must be established. However, the model is lacking so far. Furthermore, it remains unknown what the landscape looks like for deep networks of binary synapses, which plays a key role in robust and energy efficient neuromorphic computation. Here, we propose a statistical mechanics framework by directly building a least structured model of the high-dimensional weight space, considering realistic structured data, stochastic gradient descent training, and the computational depth of neural networks. We also consider whether the number of network parameters outnumbers the number of supplied training data, namely, over- or under-parametrization. Our least structured model reveals that the weight spaces of the under-parametrization and over-parameterization cases belong to the same class, in the sense that these weight spaces are well-connected without any hierarchical clustering structure. In contrast, the shallow-network has a broken weight space, characterized by a discontinuous phase transition, thereby clarifying the benefit of depth in deep learning from the angle of high dimensional geometry. Our effective model also reveals that inside a deep network, there exists a liquid-like central part of the architecture in the sense that the weights in this part behave as randomly as possible, providing algorithmic implications. Our data-driven model thus provides a statistical mechanics insight about why deep learning is unreasonably effective in terms of the high-dimensional weight space, and how deep networks are different from shallow ones.
△ Less
Submitted 28 July, 2021; v1 submitted 16 July, 2020;
originally announced July 2020.
-
Generalized bioinspired approach to a daytime radiative cooling "skin"
Authors:
Meng Yang,
Weizhi Zou,
**g Guo,
Zhenchao Qian,
Heng Luo,
Shijia Yang,
Ning Zhao,
Lorenzo Pattelli,
Jian Xu,
Diederik S. Wiersma
Abstract:
Energy-saving cooling materials with strong operability are desirable towards sustainable thermal management. Inspired by the cooperative thermo-optical effect in fur of polar bear, we develop a flexible and reusable cooling skin via laminating a polydimethylsiloxane film with a highly-scattering polyethylene aerogel. Owing to its high porosity of 97.9% and tailored pore size of 3.8 +- 1.4 microme…
▽ More
Energy-saving cooling materials with strong operability are desirable towards sustainable thermal management. Inspired by the cooperative thermo-optical effect in fur of polar bear, we develop a flexible and reusable cooling skin via laminating a polydimethylsiloxane film with a highly-scattering polyethylene aerogel. Owing to its high porosity of 97.9% and tailored pore size of 3.8 +- 1.4 micrometers, superior solar reflectance of 0.96 and high transparency to irradiated thermal energy of 0.8 can be achieved at a thickness of 2.7 mm. Combined with low thermal conductivity of 0.032 W/m/K of the aerogel, the cooling skin exerts midday sub-ambient temperature drops of 5-6 degrees in a metropolitan environment, with an estimated limit of 14 degrees under ideal service conditions. We envision that this generalized bilayer approach will construct a bridge from night-time to daytime radiative cooling and pave the way for economical, scalable, flexible and reusable cooling materials.
△ Less
Submitted 18 May, 2020;
originally announced May 2020.
-
Tunable electronic structure and stoichiometry dependent disorder in Nanostructured VO$_x$ films
Authors:
A. Delia,
S. J. Rezvani,
N. Zema,
F. Zuccaro,
M. Fanetti,
Blaz Belec,
B. W. Li,
C. W. Zou,
C. Spezzani,
M. Sacchi,
A. Marcelli,
M. Coreno
Abstract:
We present and discuss an original method to synthesize disordered Nanostructured (NS) VO$_x$ films with controlled stoichiometry and tunable electronic structures. In these NS films, the original lattice symmetry of the bulk vanadium oxides is broken and atoms are arranged in a highly disordered structure . The stoichiometry-dependent disorder as a function of the oxygen concentration has been ch…
▽ More
We present and discuss an original method to synthesize disordered Nanostructured (NS) VO$_x$ films with controlled stoichiometry and tunable electronic structures. In these NS films, the original lattice symmetry of the bulk vanadium oxides is broken and atoms are arranged in a highly disordered structure . The stoichiometry-dependent disorder as a function of the oxygen concentration has been characterized by in-situ X-ray Absorption Near-Edge Structure (XANES) spectroscopy identifying the spectroscopic fingerprints. Results show structural rearrangements that deviate from the octahedral symmetry with different coexisting disordered phases. The modulation of the electronic structure of the NS films based on the resulted stoichiometry and the quantum confinement in the NS particles are also discussed. We demonstrate the possibility to modulate the electronic structure of VO$_x$ NS films accessing new disordered atomic configurations with a controlled stoichiometry that provides an extraordinary opportunity to match a wide number of technological applications.
△ Less
Submitted 15 April, 2020;
originally announced April 2020.
-
Strain induced orbital dynamics across the Metal Insulator transition in thin VO2/TiO2(001) films
Authors:
A. D'Elia,
S. J. Rezvani,
A. Cossaro,
M. Stredansky,
C. Grazioli,
B. W. Li,
C. W. Zou,
M. Coreno,
A. Marcelli
Abstract:
VO2 is a strongly correlated material, which undergoes a reversible metal insulator transition (MIT) coupled to a structural phase transition upon heating (T= 67° C). Since its discovery the nature of the insulating state has long been debated and different solid-state mechanisms have been proposed to explain its nature: Mott-Hubbard correlation, Peierls distortion or a combination of both. Moreov…
▽ More
VO2 is a strongly correlated material, which undergoes a reversible metal insulator transition (MIT) coupled to a structural phase transition upon heating (T= 67° C). Since its discovery the nature of the insulating state has long been debated and different solid-state mechanisms have been proposed to explain its nature: Mott-Hubbard correlation, Peierls distortion or a combination of both. Moreover, still now there is a lack of consensus on the interplay between the different degrees of freedom: charge, lattice, orbital and how they contribute to the MIT. In this manuscript we will investigate across the MIT the orbital evolution induced by a tensile strain applied to thin VO2 films. The strained films allowed to study the interplay between orbital and lattice degrees of freedom and to clarify MIT properties.
△ Less
Submitted 12 January, 2020;
originally announced January 2020.
-
Biomimetic Polymer Film with Brilliant Brightness Using a One-Step Water Vapor-Induced Phase Separation Method
Authors:
Weizhi Zou,
Lorenzo Pattelli,
**g Guo,
Shijia Yang,
Meng Yang,
Ning Zhao,
Jian Xu,
Diederik S. Wiersma
Abstract:
The scales of the white Cyphochilus beetles are endowed with unusual whiteness arising from the exceptional scattering efficiency of their disordered ultrastructure optimized through millions of years of evolution. Here, a simple, one-step method based on water vapor-induced phase separation (VIPS) is developed to prepare ultra-thin polystyrene (PS) films with similar microstructure and comparable…
▽ More
The scales of the white Cyphochilus beetles are endowed with unusual whiteness arising from the exceptional scattering efficiency of their disordered ultrastructure optimized through millions of years of evolution. Here, a simple, one-step method based on water vapor-induced phase separation (VIPS) is developed to prepare ultra-thin polystyrene (PS) films with similar microstructure and comparable optical performance. A typical biomimetic 3.5 um PS film exhibits a diffuse reflectance of 61% at 500 nm, which translates into a transport mean free path below 1 um. A complete optical characterization through Monte Carlo simulations reveals how such scattering performance arises from the scattering coefficient and scattering anisotropy, whose interplay provides insight into the morphological properties of the material. The potential of bright-white coatings as smart sensors or wearable devices is highlighted using a treated ultra-thin film as a real-time sensor for human exhalation.
△ Less
Submitted 11 June, 2019;
originally announced June 2019.
-
Giant exchange bias and ferromagnetism in the CoO shell of Co/CoO-MgO core-shell nanoparticles
Authors:
C. N. Ge,
Xiangang Wan,
E. Pellegrin,
Z. Hu,
W. Q. Zou,
Y. W. Du
Abstract:
Using magnetron sputtering, we produced a series of Co/CoO-MgO nanoparticles on Si(100) substrates. High-resolution transmission electron microscopy (HRTEM) image shows that small isolated Co-clusters (core) covered with CoO (shells) with a size of a few nm embedded in a MgO matrix. Resistivity as a function of Co atomic ratio exhibits a distinct percolation threshold with a sharp decrease around…
▽ More
Using magnetron sputtering, we produced a series of Co/CoO-MgO nanoparticles on Si(100) substrates. High-resolution transmission electron microscopy (HRTEM) image shows that small isolated Co-clusters (core) covered with CoO (shells) with a size of a few nm embedded in a MgO matrix. Resistivity as a function of Co atomic ratio exhibits a distinct percolation threshold with a sharp decrease around 69% Co content. Across the threshold, the resistivity drops about 7 orders of magnitude. For a sample at this percolation critical threshold, we have observed a giant exchange bias field HE=2460 Oe at T= 2K, and using soft x-ray magnetic circular dichroism at the Co-L2,3 edge, we have detected a ferromagnetic (FM) signal originating from the antiferromagnetic CoO shell. Moreover, decreasing the Mg-impurities will reduce the FM signal from CoO shell (namely the uncompensated spin density) and the size of HE, thus directly support the uncompensated spin model.
△ Less
Submitted 9 October, 2012;
originally announced October 2012.
-
Low-temperature specific heat and thermal conductivity of glycerol
Authors:
C. Talon,
Q. W. Zou,
M. A. Ramos,
R. Villar,
S. Vieira
Abstract:
We have measured the thermal conductivity of glassy glycerol between 1.5 K and 100 K, as well as the specific heat of both glassy and crystalline phases of glycerol between 0.5 K and 25 K. We discuss both low-temperature properties of this typical molecular glass in terms of the soft-potential model. Our finding of an excellent agreement between its predictions and experimental data for these tw…
▽ More
We have measured the thermal conductivity of glassy glycerol between 1.5 K and 100 K, as well as the specific heat of both glassy and crystalline phases of glycerol between 0.5 K and 25 K. We discuss both low-temperature properties of this typical molecular glass in terms of the soft-potential model. Our finding of an excellent agreement between its predictions and experimental data for these two independent measurements constitutes a robust proof of the capabilities of the soft-potential model to account for the low-temperature properties of glasses in a wide temperature range.
△ Less
Submitted 30 October, 2001;
originally announced October 2001.
