arXiv subjects

Yuan Gao

Publications and source records attributed to Yuan Gao.

At least 487 records · Page 27Linked to original sources

arXiv2020

Non-local SPDE limits of spatially-correlated-noise driven spin systems derived to sample a canonical distribution

We study the macroscopic behavior of a stochastic spin ensemble driven by a discrete Markov jump process motivated by the Metropolis-Hastings algorithm where the proposal is made with spatially correlated (colored) noise, and hence fails to be symmetric. However, we demonstrate a scenario where the failure of proposal symmetry is a higher order effect. Hence, from these microscopic dynamics we derive as a limit as the proposal size goes to zero and the number of spins to infinity, a non-local stochastic version of the harmonic map heat flow (or overdamped Landau-Lipshitz equation). The equation is both mathematically well-posed and samples the canonical/Gibbs distribution related to the kinetic energy. The failure of proposal symmetry due to interaction between the confining geometry of the spin system and the colored noise is in contrast to the uncorrelated, white-noise, driven system. Specifically, the choice of projection of the noise to conserve the magnitude of the spins is crucial to maintaining the proper equilibrium distribution. Numerical simulations are included to verify convergence properties and demonstrate the dynamics.

cond-mat.stat-mech↗

arXiv2020

Graph-PCNN: Two Stage Human Pose Estimation with Graph Pose Refinement

Recently, most of the state-of-the-art human pose estimation methods are based on heatmap regression. The final coordinates of keypoints are obtained by decoding heatmap directly. In this paper, we aim to find a better approach to get more accurate localization results. We mainly put forward two suggestions for improvement: 1) different features and methods should be applied for rough and accurate localization, 2) relationship between keypoints should be considered. Specifically, we propose a two-stage graph-based and model-agnostic framework, called Graph-PCNN, with a localization subnet and a graph pose refinement module added onto the original heatmap regression network. In the first stage, heatmap regression network is applied to obtain a rough localization result, and a set of proposal keypoints, called guided points, are sampled. In the second stage, for each guided point, different visual feature is extracted by the localization subnet. The relationship between guided points is explored by the graph pose refinement module to get more accurate localization results. Experiments show that Graph-PCNN can be used in various backbones to boost the performance by a large margin. Without bells and whistles, our best model can achieve a new state-of-the-art 76.8% AP on COCO test-dev split.

cs.CV↗

arXiv2020

Automatic Differentiation for Second Renormalization of Tensor Networks

Tensor renormalization group (TRG) constitutes an important methodology for accurate simulations of strongly correlated lattice models. Facilitated by the automatic differentiation technique widely used in deep learning, we propose a uniform framework of differentiable TRG ($\partial$TRG) that can be applied to improve various TRG methods, in an automatic fashion. Essentially, $\partial$TRG systematically extends the concept of second renormalization [PRL 103, 160601 (2009)] where the tensor environment is computed recursively in the backward iteration, in the sense that given the forward process of TRG, $\partial$TRG automatically finds the gradient through backpropagation, with which one can deeply "train" the tensor networks. We benchmark $\partial$TRG in solving the square-lattice Ising model, and demonstrate its power by simulating one- and two-dimensional quantum systems at finite temperature. The deep optimization as well as GPU acceleration renders $\partial$TRG manybody simulations with high efficiency and accuracy.

cond-mat.str-el↗

arXiv2020

Increasing Iterate Averaging for Solving Saddle-Point Problems

Many problems in machine learning and game theory can be formulated as saddle-point problems, for which various first-order methods have been developed and proven efficient in practice. Under the general convex-concave assumption, most first-order methods only guarantee an ergodic convergence rate, that is, the uniform averages of the iterates converge at a $O(1/T)$ rate in terms of the saddle-point residual. However, numerically, the iterates themselves can often converge much faster than the uniform averages. This observation motivates increasing averaging schemes that put more weight on later iterates, in contrast to the usual uniform averaging. We show that such increasing averaging schemes, applied to various first-order methods, are able to preserve the $O(1/T)$ convergence rate with no additional assumptions or computational overhead. Extensive numerical experiments on zero-sum game solving, market equilibrium computation and image denoising demonstrate the effectiveness of the proposed schemes. In particular, the increasing averages consistently outperform the uniform averages in all test problems by orders of magnitude. When solving matrix and extensive-form games, increasing averages consistently outperform the last iterates as well. For matrix games, a first-order method equipped with increasing averaging outperforms the highly competitive CFR$^+$ algorithm.

cs.LG↗

arXiv2020

Inbetweening auto-animation via Fokker-Planck dynamics and thresholding

We propose an equilibrium-driven deformation algorithm (EDDA) to simulate the inbetweening transformations starting from an initial image to an equilibrium image, which covers images varying from a greyscale type to a colorful type on plane or manifold. The algorithm is based on Fokker-Planck dynamics on manifold, which automatically cooperates positivity, unconditional stability, mass conservation law, exponentially convergence and also the manifold structure suggested by dataset. The thresholding scheme is adapted for the sharp interface dynamics and is used to achieve the finite time convergence. Using EDDA, three challenging examples, (I) facial aging process, (II) coronavirus disease 2019 (COVID-19) invading/treatment process, and (III) continental evolution process are conducted efficiently.

math.NA↗

arXiv2020

Intra-Ensemble in Neural Networks

Improving model performance is always the key problem in machine learning including deep learning. However, stand-alone neural networks always suffer from marginal effect when stacking more layers. At the same time, ensemble is an useful technique to further enhance model performance. Nevertheless, training several independent deep neural networks for ensemble costs multiple resources. If so, is it possible to utilize ensemble in only one neural network? In this work, we propose Intra-Ensemble, an end-to-end ensemble strategy with stochastic channel recombination operations to train several sub-networks simultaneously within one neural network. Additional parameter size is marginal since the majority of parameters are mutually shared. Meanwhile, stochastic channel recombination significantly increases the diversity of sub-networks, which finally enhances ensemble performance. Extensive experiments and ablation studies prove the applicability of intra-ensemble on various kinds of datasets and network architectures.

cs.CV↗

arXiv2020

Generation of Chinese-characters-like modes with transverse mode locked lasers

Spontaneous transverse mode locking to generate optical vortex arrays is already widely studied for the frequency degenerated transverse modes. While, the mode locking between transverse modes in different orders (non-frequency degenerated families) to achieve spatial stationary beam patterns is rarely reported. The theory of transverse mode locking is discussed. Concepts of general transverse mode locking (G-TML) and restricted transverse mode locking (R-TML) are mentioned. It's experimentally shown that R-TML is possible in microchip cavities with high nonlinearity. More interestingly, some Chinese characters like laser modes can be generated by microchip lasers with this special transverse mode locking effect. And various experimental results and corresponding simulations are presented.

nlin.PS↗

arXiv2020

Unity: A General Platform for Intelligent Agents

Recent advances in artificial intelligence have been driven by the presence of increasingly realistic and complex simulated environments. However, many of the existing environments provide either unrealistic visuals, inaccurate physics, low task complexity, restricted agent perspective, or a limited capacity for interaction among artificial agents. Furthermore, many platforms lack the ability to flexibly configure the simulation, making the simulated environment a black-box from the perspective of the learning system. In this work, we propose a novel taxonomy of existing simulation platforms and discuss the highest level class of general platforms which enable the development of learning environments that are rich in visual, physical, task, and social complexity. We argue that modern game engines are uniquely suited to act as general platforms and as a case study examine the Unity engine and open source Unity ML-Agents Toolkit. We then survey the research enabled by Unity and the Unity ML-Agents Toolkit, discussing the kinds of research a flexible, interactive and easily configurable general platform can facilitate.

cs.LG↗

arXiv2020

Weakly Supervised Deep Learning for COVID-19 Infection Detection and Classification from CT Images

An outbreak of a novel coronavirus disease (i.e., COVID-19) has been recorded in Wuhan, China since late December 2019, which subsequently became pandemic around the world. Although COVID-19 is an acutely treated disease, it can also be fatal with a risk of fatality of 4.03% in China and the highest of 13.04% in Algeria and 12.67% Italy (as of 8th April 2020). The onset of serious illness may result in death as a consequence of substantial alveolar damage and progressive respiratory failure. Although laboratory testing, e.g., using reverse transcription polymerase chain reaction (RT-PCR), is the golden standard for clinical diagnosis, the tests may produce false negatives. Moreover, under the pandemic situation, shortage of RT-PCR testing resources may also delay the following clinical decision and treatment. Under such circumstances, chest CT imaging has become a valuable tool for both diagnosis and prognosis of COVID-19 patients. In this study, we propose a weakly supervised deep learning strategy for detecting and classifying COVID-19 infection from CT images. The proposed method can minimise the requirements of manual labelling of CT images but still be able to obtain accurate infection detection and distinguish COVID-19 from non-COVID-19 cases. Based on the promising results obtained qualitatively and quantitatively, we can envisage a wide deployment of our developed technique in large-scale clinical studies.

eess.IV↗

arXiv2020

Limiting Behaviors of High Dimensional Stochastic Spin Ensembles

Lattice spin models in statistical physics are used to understand magnetism. Their Hamiltonians are a discrete form of a version of a Dirichlet energy, signifying a relationship to the Harmonic map heat flow equation. The Gibbs distribution, defined with this Hamiltonian, is used in the Metropolis-Hastings (M-H) algorithm to generate dynamics tending towards an equilibrium state. In the limiting situation when the inverse temperature is large, we establish the relationship between the discrete M-H dynamics and the continuous Harmonic map heat flow associated with the Hamiltonian. We show the convergence of the M-H dynamics to the Harmonic map heat flow equation in two steps: First, with fixed lattice size and proper choice of proposal size in one M-H step, the M-H dynamics acts as gradient descent and will be shown to converge to a system of Langevin stochastic differential equations (SDE). Second, with proper scaling of the inverse temperature in the Gibbs distribution and taking the lattice size to infinity, it will be shown that this SDE system converges to the deterministic Harmonic map heat flow equation. Our results are not unexpected, but show remarkable connections between the M-H steps and the SDE Stratonovich formulation, as well as reveal trajectory-wise out of equilibrium dynamics to be related to a canonical PDE system with geometric constraints.

math.PR↗

arXiv2020

MTL-NAS: Task-Agnostic Neural Architecture Search towards General-Purpose Multi-Task Learning

We propose to incorporate neural architecture search (NAS) into general-purpose multi-task learning (GP-MTL). Existing NAS methods typically define different search spaces according to different tasks. In order to adapt to different task combinations (i.e., task sets), we disentangle the GP-MTL networks into single-task backbones (optionally encode the task priors), and a hierarchical and layerwise features sharing/fusing scheme across them. This enables us to design a novel and general task-agnostic search space, which inserts cross-task edges (i.e., feature fusion connections) into fixed single-task network backbones. Moreover, we also propose a novel single-shot gradient-based search algorithm that closes the performance gap between the searched architectures and the final evaluation architecture. This is realized with a minimum entropy regularization on the architecture weights during the search phase, which makes the architecture weights converge to near-discrete values and therefore achieves a single model. As a result, our searched model can be directly used for evaluation without (re-)training from scratch. We perform extensive experiments using different single-task backbones on various task sets, demonstrating the promising performance obtained by exploiting the hierarchical and layerwise features, as well as the desirable generalizability to different i) task sets and ii) single-task backbones. The code of our paper is available at https://github.com/bhpfelix/MTLNAS.

cs.LG↗

arXiv2020

Stochastic Flows and Geometric Optimization on the Orthogonal Group

We present a new class of stochastic, geometrically-driven optimization algorithms on the orthogonal group $O(d)$ and naturally reductive homogeneous manifolds obtained from the action of the rotation group $SO(d)$. We theoretically and experimentally demonstrate that our methods can be applied in various fields of machine learning including deep, convolutional and recurrent neural networks, reinforcement learning, normalizing flows and metric learning. We show an intriguing connection between efficient stochastic optimization on the orthogonal group and graph theory (e.g. matching problem, partition functions over graphs, graph-coloring). We leverage the theory of Lie groups and provide theoretical results for the designed class of algorithms. We demonstrate broad applicability of our methods by showing strong performance on the seemingly unrelated tasks of learning world models to obtain stable policies for the most difficult $\mathrm{Humanoid}$ agent from $\mathrm{OpenAI}$ $\mathrm{Gym}$ and improving convolutional neural networks.

cs.LG↗

arXiv2020

Existence and uniqueness of bounded stable solutions to the Peierls-Nabarro model for curved dislocations

We study the well-posedness of the vector-field Peierls-Nabarro model for curved dislocations with a double well potential and a bi-states limit at far field. Using the Dirichlet to Neumann map, the 3D Peierls-Nabarro model is reduced to a nonlocal scalar Ginzburg-Landau equation. We derive an integral formulation of the nonlocal operator, whose kernel is anisotropic and positive when Poisson's ratio $ν\in(-\frac12, \frac13)$. We then prove that any bounded stable solutions to this nonlocal scalar Ginzburg-Landau equation has a 1D profile, which corresponds to the PDE version of flatness result for minimal surfaces with anisotropic nonlocal perimeter. Based on this, we finally obtain that steady states to the nonlocal scalar equation, as well as the original Peierls-Nabarro model, can be characterized as a one-parameter family of straight dislocation solutions to a rescaled 1D Ginzburg-Landau equation with the half Laplacian.

math.AP↗

arXiv2020

Self-Supervised Light Field Reconstruction Using Shearlet Transform and Cycle Consistency

The image-based rendering approach using Shearlet Transform (ST) is one of the state-of-the-art Densely-Sampled Light Field (DSLF) reconstruction methods. It reconstructs Epipolar-Plane Images (EPIs) in image domain via an iterative regularization algorithm restoring their coefficients in shearlet domain. Consequently, the ST method tends to be slow because of the time spent on domain transformations for dozens of iterations. To overcome this limitation, this letter proposes a novel self-supervised DSLF reconstruction method, CycleST, which applies ST and cycle consistency to DSLF reconstruction. Specifically, CycleST is composed of an encoder-decoder network and a residual learning strategy that restore the shearlet coefficients of densely-sampled EPIs using EPI reconstruction and cycle consistency losses. Besides, CycleST is a self-supervised approach that can be trained solely on Sparsely-Sampled Light Fields (SSLFs) with small disparity ranges ($\leqslant$ 8 pixels). Experimental results of DSLF reconstruction on SSLFs with large disparity ranges (16 - 32 pixels) from two challenging real-world light field datasets demonstrate the effectiveness and efficiency of the proposed CycleST method. Furthermore, CycleST achieves ~ 9x speedup over ST, at least.

eess.SP↗

arXiv2020

DRST: Deep Residual Shearlet Transform for Densely Sampled Light Field Reconstruction

The Image-Based Rendering (IBR) approach using Shearlet Transform (ST) is one of the most effective methods for Densely-Sampled Light Field (DSLF) reconstruction. The ST-based DSLF reconstruction typically relies on an iterative thresholding algorithm for Epipolar-Plane Image (EPI) sparse regularization in shearlet domain, involving dozens of transformations between image domain and shearlet domain, which are in general time-consuming. To overcome this limitation, a novel learning-based ST approach, referred to as Deep Residual Shearlet Transform (DRST), is proposed in this paper. Specifically, for an input sparsely-sampled EPI, DRST employs a deep fully Convolutional Neural Network (CNN) to predict the residuals of the shearlet coefficients in shearlet domain in order to reconstruct a densely-sampled EPI in image domain. The DRST network is trained on synthetic Sparsely-Sampled Light Field (SSLF) data only by leveraging elaborately-designed masks. Experimental results on three challenging real-world light field evaluation datasets with varying moderate disparity ranges (8 - 16 pixels) demonstrate the superiority of the proposed learning-based DRST approach over the non-learning-based ST method for DSLF reconstruction. Moreover, DRST provides a 2.4x speedup over ST, at least.

cs.MM↗

arXiv2020

Application of Deep Q-Network in Portfolio Management

Machine Learning algorithms and Neural Networks are widely applied to many different areas such as stock market prediction, face recognition and population analysis. This paper will introduce a strategy based on the classic Deep Reinforcement Learning algorithm, Deep Q-Network, for portfolio management in stock market. It is a type of deep neural network which is optimized by Q Learning. To make the DQN adapt to financial market, we first discretize the action space which is defined as the weight of portfolio in different assets so that portfolio management becomes a problem that Deep Q-Network can solve. Next, we combine the Convolutional Neural Network and dueling Q-net to enhance the recognition ability of the algorithm. Experimentally, we chose five lowrelevant American stocks to test the model. The result demonstrates that the DQN based strategy outperforms the ten other traditional strategies. The profit of DQN algorithm is 30% more than the profit of other strategies. Moreover, the Sharpe ratio associated with Max Drawdown demonstrates that the risk of policy made with DQN is the lowest.

q-fin.PM↗

arXiv2020

Learning Implicit Generative Models with Theoretical Guarantees

We propose a \textbf{uni}fied \textbf{f}ramework for \textbf{i}mplicit \textbf{ge}nerative \textbf{m}odeling (UnifiGem) with theoretical guarantees by integrating approaches from optimal transport, numerical ODE, density-ratio (density-difference) estimation and deep neural networks. First, the problem of implicit generative learning is formulated as that of finding the optimal transport map between the reference distribution and the target distribution, which is characterized by a totally nonlinear Monge-Ampère equation. Interpreting the infinitesimal linearization of the Monge-Ampère equation from the perspective of gradient flows in measure spaces leads to the continuity equation or the McKean-Vlasov equation. We then solve the McKean-Vlasov equation numerically using the forward Euler iteration, where the forward Euler map depends on the density ratio (density difference) between the distribution at current iteration and the underlying target distribution. We further estimate the density ratio (density difference) via deep density-ratio (density-difference) fitting and derive explicit upper bounds on the estimation error. Experimental results on both synthetic datasets and real benchmark datasets support our theoretical findings and demonstrate the effectiveness of UnifiGem.

cs.LG↗

arXiv2019

Automated Testing for Deep Learning Systems with Differential Behavior Criteria

In this work, we conducted a study on building an automated testing system for deep learning systems based on differential behavior criteria. The automated testing goals were achieved by jointly optimizing two objective functions: maximizing differential behaviors from models under testing and maximizing neuron coverage. By observing differential behaviors from three pre-trained models during each testing iteration, the input image that triggered erroneous feedback was registered as a corner-case. The generated corner-cases can be used to examine the robustness of DNNs and consequently improve model accuracy. A project called DeepXplore was also used as a baseline model. After we fully implemented and optimized the baseline system, we explored its application as an augmenting training dataset with newly generated corner cases. With the GTRSB dataset, by retraining the model based on automated generated corner cases, the accuracy of three generic models increased by 259.2%, 53.6%, and 58.3%, respectively. Further, to extend the capability of automated testing, we explored other approaches based on differential behavior criteria to generate photo-realistic images for deep learning systems. One approach was to apply various transformations to the seed images for the deep learning framework. The other approach was to utilize the Generative Adversarial Networks (GAN) technique, which was implemented on MNIST and Driving datasets. The style transferring capability has been observed very effective in adding additional visual effects, replacing image elements, and style-shifting (virtual image to real images). The GAN-based testing sample generation system was shown to be the next frontier for automated testing for deep learning systems.

cs.LG↗