arXiv subjects

Xin Guo

Publications and source records attributed to Xin Guo.

At least 145 records · Page 8Linked to original sources

arXiv2022

Out-of-distribution Generalization via Partial Feature Decorrelation

Most deep-learning-based image classification methods assume that all samples are generated under an independent and identically distributed (IID) setting. However, out-of-distribution (OOD) generalization is more common in practice, which means an agnostic context distribution shift between training and testing environments. To address this problem, we present a novel Partial Feature Decorrelation Learning (PFDL) algorithm, which jointly optimizes a feature decomposition network and the target image classification model. The feature decomposition network decomposes feature embeddings into the independent and the correlated parts such that the correlations between features will be highlighted. Then, the correlated features help learn a stable feature representation by decorrelating the highlighted correlations while optimizing the image classification model. We verify the correlation modeling ability of the feature decomposition network on a synthetic dataset. The experiments on real-world datasets demonstrate that our method can improve the backbone model's accuracy on OOD image classification datasets.

cs.LG↗

arXiv2022

Mean-Field Multi-Agent Reinforcement Learning: A Decentralized Network Approach

One of the challenges for multi-agent reinforcement learning (MARL) is designing efficient learning algorithms for a large system in which each agent has only limited or partial information of the entire system. While exciting progress has been made to analyze decentralized MARL with the network of agents for social networks and team video games, little is known theoretically for decentralized MARL with the network of states for modeling self-driving vehicles, ride-sharing, and data and traffic routing. This paper proposes a framework of localized training and decentralized execution to study MARL with network of states. Localized training means that agents only need to collect local information in their neighboring states during the training phase; decentralized execution implies that agents can execute afterwards the learned decentralized policies, which depend only on agents' current states. The theoretical analysis consists of three key components: the first is the reformulation of the MARL system as a networked Markov decision process with teams of agents, enabling updating the associated team Q-function in a localized fashion; the second is the Bellman equation for the value function and the appropriate Q-function on the probability measure space; and the third is the exponential decay property of the team Q-function, facilitating its approximation with efficient sample efficiency and controllable error. The theoretical analysis paves the way for a new algorithm LTDE-Neural-AC, where the actor-critic approach with over-parameterized neural networks is proposed. The convergence and sample complexity is established and shown to be scalable with respect to the sizes of both agents and states. To the best of our knowledge, this is the first neural network based MARL algorithm with network structure and provably convergence guarantee.

cs.LG↗

arXiv2022

SDE approximations of GANs training and its long-run behavior

This paper analyzes the training process of GANs via stochastic differential equations (SDEs). It first establishes SDE approximations for the training of GANs under stochastic gradient algorithms, with precise error bound analysis. It then describes the long-run behavior of GANs training via the invariant measures of its SDE approximations under proper conditions. This work builds theoretical foundation for GANs training and provides analytical tools to study its evolution and stability.

cs.LG↗

arXiv2022

Meta-learning with GANs for anomaly detection, with deployment in high-speed rail inspection system

Anomaly detection has been an active research area with a wide range of potential applications. Key challenges for anomaly detection in the AI era with big data include lack of prior knowledge of potential anomaly types, highly complex and noisy background in input data, scarce abnormal samples, and imbalanced training dataset. In this work, we propose a meta-learning framework for anomaly detection to deal with these issues. Within this framework, we incorporate the idea of generative adversarial networks (GANs) with appropriate choices of loss functions including structural similarity index measure (SSIM). Experiments with limited labeled data for high-speed rail inspection demonstrate that our meta-learning framework is sharp and robust in identifying anomalies. Our framework has been deployed in five high-speed railways of China since 2021: it has reduced more than 99.7% workload and saved 96.7% inspection time.

cs.LG↗

arXiv2022

Atomically smooth single-crystalline platform for low-loss plasmonic nanocavities

Nanoparticle-on-mirror plasmonic nanocavities, capable of extreme optical confinement and enhancement, have triggered state-of-the-art progress in nanophotonics and development of applications in enhanced spectroscopies and molecular detection. However, the optical quality factor and thus performance of these nanoconstructs are undermined by the granular polycrystalline metal films used as a mirror. Here, we report an atomically smooth single-crystalline platform for low-loss nanocavities using chemically-synthesized gold microflakes as a mirror. Nanocavities constructed using gold nanorods on such microflakes exhibit a rich structure of plasmonic modes, which are highly sensitive to the thickness of optically-thin (down to ~15 nm) microflakes. The atomically smooth single-crystalline microflakes endow nanocavities with significantly improved quality factor (~2 times) and scattering intensity (~3 times) compared with their counterparts based on deposited films. The developed low-loss nanocavities further allow for the integration with a mature platform of fiber optics, opening opportunities for realizing nanocavity-based miniaturized photonic devices with high performance.

physics.optics↗

arXiv2022

Transfer Learning for Retinal Vascular Disease Detection: A Pilot Study with Diabetic Retinopathy and Retinopathy of Prematurity

Retinal vascular diseases affect the well-being of human body and sometimes provide vital signs of otherwise undetected bodily damage. Recently, deep learning techniques have been successfully applied for detection of diabetic retinopathy (DR). The main obstacle of applying deep learning techniques to detect most other retinal vascular diseases is the limited amount of data available. In this paper, we propose a transfer learning technique that aims to utilize the feature similarities for detecting retinal vascular diseases. We choose the well-studied DR detection as a source task and identify the early detection of retinopathy of prematurity (ROP) as the target task. Our experimental results demonstrate that our DR-pretrained approach dominates in all metrics the conventional ImageNet-pretrained transfer learning approach, currently adopted in medical image analysis. Moreover, our approach is more robust with respect to the stochasticity in the training process and with respect to reduced training samples. This study suggests the potential of our proposed transfer learning approach for a broad range of retinal vascular diseases or pathologies, where data is limited.

cs.LG↗

arXiv2021

Convergence of GANs Training: A Game and Stochastic Control Methodology

Training generative adversarial networks (GANs) is known to be difficult, especially for financial time series. This paper first analyzes the well-posedness problem in GANs minimax games and the convexity issue in GANs objective functions. It then proposes a stochastic control framework for hyper-parameters tuning in GANs training. The weak form of dynamic programming principle and the uniqueness and the existence of the value function in the viscosity sense for the corresponding minimax game are established. In particular, explicit forms for the optimal adaptive learning rate and batch size are derived and are shown to depend on the convexity of the objective function, revealing a relation between improper choices of learning rate and explosion in GANs training. Finally, empirical studies demonstrate that training algorithms incorporating this adaptive control approach outperform the standard ADAM method in terms of convergence and robustness. From GANs training perspective, the analysis in this paper provides analytical support for the popular practice of ``clipping'', and suggests that the convexity and well-posedness issues in GANs may be tackled through appropriate choices of hyper-parameters.

stat.ML↗

arXiv2021

SiamTrans: Zero-Shot Multi-Frame Image Restoration with Pre-Trained Siamese Transformers

We propose a novel zero-shot multi-frame image restoration method for removing unwanted obstruction elements (such as rains, snow, and moire patterns) that vary in successive frames. It has three stages: transformer pre-training, zero-shot restoration, and hard patch refinement. Using the pre-trained transformers, our model is able to tell the motion difference between the true image information and the obstructing elements. For zero-shot image restoration, we design a novel model, termed SiamTrans, which is constructed by Siamese transformers, encoders, and decoders. Each transformer has a temporal attention layer and several self-attention layers, to capture both temporal and spatial information of multiple frames. Only pre-trained (self-supervised) on the denoising task, SiamTrans is tested on three different low-level vision tasks (deraining, demoireing, and desnowing). Compared with related methods, ours achieves the best performances, even outperforming those with supervised learning.

cs.CV↗

arXiv2021

Transaction Cost Analytics for Corporate Bonds

The electronic platform has been increasingly popular for executing large corporate bond orders by asset managers, who in turn have to assess the quality of their executions via Transaction Cost Analysis (TCA). One of the challenges in TCA is to build a realistic benchmark for the expected transaction cost and to characterize the price impact of each individual trade with given bond characteristics and market conditions. Taking the viewpoint of retail investors, this paper presents an analytical methodology for TCA of corporate bond trading. Our analysis is based on the TRACE Enhanced dataset; and starts with estimating the initiator of a bond transaction, followed by estimating the bid-ask spread and the mid-price dynamics. With these estimations, the first part of our study is to identify key features for corporate bonds and to compute the expected average trading cost. This part is on the time scale of weekly transactions, and is by applying and comparing several regularized regression models. The second part of our study is using the estimated mid-price dynamics to investigate the amplitude of its price impact and the decay pattern of individual bond transaction. This part is on the time scale of each transaction of liquid corporate bonds, and is by applying a transient impact model to estimate the price impact kernel using a non-parametric method. Our benchmark model allows for identifying abnormal transactions and for enhancing counter-party selections. A key discovery of our study is the price impact asymmetry between customer-buy orders and consumer-sell orders.

q-fin.ST↗

arXiv2021

Entropy Regularization for Mean Field Games with Learning

Entropy regularization has been extensively adopted to improve the efficiency, the stability, and the convergence of algorithms in reinforcement learning. This paper analyzes both quantitatively and qualitatively the impact of entropy regularization for Mean Field Game (MFG) with learning in a finite time horizon. Our study provides a theoretical justification that entropy regularization yields time-dependent policies and, furthermore, helps stabilizing and accelerating convergence to the game equilibrium. In addition, this study leads to a policy-gradient algorithm for exploration in MFG. Under this algorithm, agents are able to learn the optimal exploration scheduling, with stable and fast convergence to the game equilibrium.

math.OC↗

arXiv2021

Extracting and Measuring Uncertain Biomedical Knowledge from Scientific Statements

Purpose: This study aims to develop a novel approach to extracting and measuring uncertain biomedical knowledge from scientific statements. Design/methodology/approach: Taking cardiovascular research publications in China as a sample, we extracted the SPO triples as knowledge unit and the hedging/conflicting uncertainties as the knowledge context. We introduced Information Entropy and Uncertainty Rate as potential metrics to quantity the uncertainty of biomedical knowledge claims represented at different levels, such as the SPO triples (micro level), as well as the semantic type pairs (micro-level). Findings: The results indicated that while the number of scientific publications and total SPO triples showed a liner growth, the novel SPO triples occurring per year remained stable. After examining the frequency of uncertain cue words in different part of scientific statements, we found hedging words tend to appear in conclusive and purposeful sentences, whereas conflicting terms often appear in background and act as the premise (e.g., unsettled scientific issues) of the work to be investigated. Practical implications: Our approach identified major uncertain knowledge areas, such as diagnostic biomarkers, genetic characteristics, and pharmacologic therapies surrounding cardiovascular diseases in China. These areas are suggested to be prioritized in which new hypotheses need to be verified, and disputes, conflicts, as well as contradictions to be settled further.

cs.IR↗

arXiv2021

A General Framework for Lifelong Localization and Mapping in Changing Environment

The environment of most real-world scenarios such as malls and supermarkets changes at all times. A pre-built map that does not account for these changes becomes out-of-date easily. Therefore, it is necessary to have an up-to-date model of the environment to facilitate long-term operation of a robot. To this end, this paper presents a general lifelong simultaneous localization and mapping (SLAM) framework. Our framework uses a multiple session map representation, and exploits an efficient map updating strategy that includes map building, pose graph refinement and sparsification. To mitigate the unbounded increase of memory usage, we propose a map-trimming method based on the Chow-Liu maximum-mutual-information spanning tree. The proposed SLAM framework has been comprehensively validated by over a month of robot deployment in real supermarket environment. Furthermore, we release the dataset collected from the indoor and outdoor changing environment with the hope to accelerate lifelong SLAM research in the community. Our dataset is available at https://github.com/sanduan168/lifelong-SLAM-dataset.

cs.RO↗

arXiv2021

A class of stochastic games and moving free boundary problems

In this paper we propose and analyze a class of $N$-player stochastic games that include finite fuel stochastic games as a special case. We first derive sufficient conditions for the Nash equilibrium (NE) in the form of a verification theorem. The associated Quasi-Variational-Inequalities include an essential game component regarding the interactions among players, which may be interpreted as the analytical representation of the conditional optimality for NEs. The derivation of NEs involves solving first a multi-dimensional free boundary problem and then a Skorokhod problem. We call it a "moving free boundary" to highlight the difference between standard control problems and stochastic games. Finally, we present an intriguing connection between these NE strategies and controlled rank-dependent stochastic differential equations.

q-fin.MF↗

arXiv2021

The energetics of quantum vacuum friction: Field fluctuations

Quantum fluctuations can induce a friction on a neutral but polarizable particle and cause it to radiate energy even if the particle is moving in free space filled with blackbody radiation, and is not in contact with or close to any surface or other object. We explore the energetics of such a particle moving uniformly in vacuum, continuing our previous investigations of quantum friction. The intrinsic polarizability of the particle is considered to be purely real before it is dressed by radiation. The particle is then guaranteed to be in the nonequilibrium steady state (NESS), where it absorbs and emits energy at the same rate. We first calculate the quantum frictional power and force on the particle in the rest frame of the blackbody radiation from first principles, namely the Maxwell-Heaviside equations and the Lorentz force law. Then we provide a simpler method of obtaining the same quantities in the rest frame of the particle by using the principle of virtual work. The equivalence of the two approaches is illustrated. The formulas we derive for quantum vacuum frictional power and force are fully relativistic and applicable to finite temperature. In NESS, the quantum vacuum frictional force on the particle is shown to be a true drag, independent of the model for polarizability and the polarization state of the particle. Finally, we give an estimate of the quantum vacuum friction on a gold atom and comment on the feasibility of detecting such quantum vacuum frictional effects.

hep-th↗

arXiv2021

Learning Mean-Field Games

This paper presents a general mean-field game (GMFG) framework for simultaneous learning and decision-making in stochastic games with a large population. It first establishes the existence of a unique Nash Equilibrium to this GMFG, and explains that naively combining Q-learning with the fixed-point approach in classical MFGs yields unstable algorithms. It then proposes a Q-learning algorithm with Boltzmann policy (GMF-Q), with analysis of convergence property and computational complexity. The experiments on repeated Ad auction problems demonstrate that this GMF-Q algorithm is efficient and robust in terms of convergence and learning accuracy. Moreover, its performance is superior in convergence, stability, and learning ability, when compared with existing algorithms for multi-agent reinforcement learning.

math.OC↗

arXiv2021

Mean-Field Controls with Q-learning for Cooperative MARL: Convergence and Complexity Analysis

Multi-agent reinforcement learning (MARL), despite its popularity and empirical success, suffers from the curse of dimensionality. This paper builds the mathematical framework to approximate cooperative MARL by a mean-field control (MFC) approach, and shows that the approximation error is of $\mathcal{O}(\frac{1}{\sqrt{N}})$. By establishing an appropriate form of the dynamic programming principle for both the value function and the Q function, it proposes a model-free kernel-based Q-learning algorithm (MFC-K-Q), which is shown to have a linear convergence rate for the MFC problem, the first of its kind in the MARL literature. It further establishes that the convergence rate and the sample complexity of MFC-K-Q are independent of the number of agents $N$, which provides an $\mathcal{O}(\frac{1}{\sqrt{N}})$ approximation to the MARL problem with $N$ agents in the learning environment. Empirical studies for the network traffic congestion problem demonstrate that MFC-K-Q outperforms existing MARL algorithms when $N$ is large, for instance when $N>50$.

cs.LG↗

arXiv2021

Theoretical Guarantees of Fictitious Discount Algorithms for Episodic Reinforcement Learning and Global Convergence of Policy Gradient Methods

When designing algorithms for finite-time-horizon episodic reinforcement learning problems, a common approach is to introduce a fictitious discount factor and use stationary policies for approximations. Empirically, it has been shown that the fictitious discount factor helps reduce variance, and stationary policies serve to save the per-iteration computational cost. Theoretically, however, there is no existing work on convergence analysis for algorithms with this fictitious discount recipe. This paper takes the first step towards analyzing these algorithms. It focuses on two vanilla policy gradient (VPG) variants: the first being a widely used variant with discounted advantage estimations (DAE), the second with an additional fictitious discount factor in the score functions of the policy gradient estimators. Non-asymptotic convergence guarantees are established for both algorithms, and the additional discount factor is shown to reduce the bias introduced in DAE and thus improve the algorithm convergence asymptotically. A key ingredient of our analysis is to connect three settings of Markov decision processes (MDPs): the finite-time-horizon, the average reward and the discounted settings. To our best knowledge, this is the first theoretical guarantee on fictitious discount algorithms for the episodic reinforcement learning of finite-time-horizon MDPs, which also leads to the (first) global convergence of policy gradient methods for finite-time-horizon episodic reinforcement learning.

cs.LG↗

arXiv2021

Generative Adversarial Network: Some Analytical Perspectives

Ever since its debut, generative adversarial networks (GANs) have attracted tremendous amount of attention. Over the past years, different variations of GANs models have been developed and tailored to different applications in practice. Meanwhile, some issues regarding the performance and training of GANs have been noticed and investigated from various theoretical perspectives. This subchapter will start from an introduction of GANs from an analytical perspective, then move on to the training of GANs via SDE approximations and finally discuss some applications of GANs in computing high dimensional MFGs as well as tackling mathematical finance problems.

q-fin.MF↗