Search Results for author: Richard Yi Da Xu

Found 28 papers, 4 papers with code

Demystify Optimization and Generalization of Over-parameterized PAC-Bayesian Learning

no code implementations4 Feb 2022 Wei Huang, Chunrui Liu, Yilan Chen, Tianyu Liu, Richard Yi Da Xu

In addition to being a pure generalization bound analysis tool, PAC-Bayesian bound can also be incorporated into an objective function to train a probabilistic neural network, making them a powerful and relevant framework that can numerically provide a tight generalization bound for supervised learning.

Dual Behavior Regularized Reinforcement Learning

no code implementations19 Sep 2021 Chapman Siu, Jason Traish, Richard Yi Da Xu

We demonstrate the flexibility of this approach and how it can be adapted to online contexts where the environment is available to collect experiences and a variety of other contexts.

reinforcement-learning

Regularize! Don't Mix: Multi-Agent Reinforcement Learning without Explicit Centralized Structures

no code implementations19 Sep 2021 Chapman Siu, Jason Traish, Richard Yi Da Xu

We propose using regularization for Multi-Agent Reinforcement Learning rather than learning explicit cooperative structures called {\em Multi-Agent Regularized Q-learning} (MARQ).

Multi-agent Reinforcement Learning Q-Learning +1

Alleviating Mode Collapse in GAN via Diversity Penalty Module

no code implementations5 Aug 2021 Sen Pei, Richard Yi Da Xu, Shiming Xiang, Gaofeng Meng

We compare the proposed method with Unrolled GAN (Metz et al. 2016), BourGAN (Xiao, Zhong, and Zheng 2018), PacGAN (Lin et al. 2018), VEEGAN (Srivastava et al. 2017) and ALI (Dumoulin et al. 2016) on 2D synthetic dataset, and results show that the diversity penalty module can help GAN capture much more modes of the data distribution.

Data Augmentation

Towards Deepening Graph Neural Networks: A GNTK-based Optimization Perspective

no code implementations ICLR 2022 Wei Huang, Yayong Li, Weitao Du, Jie Yin, Richard Yi Da Xu, Ling Chen, Miao Zhang

Inspired by our theoretical insights on trainability, we propose Critical DropEdge, a connectivity-aware and graph-adaptive sampling method, to alleviate the exponential decay problem more fundamentally.

Resolution-invariant Person ReID Based on Feature Transformation and Self-weighted Attention

no code implementations12 Jan 2021 Ziyue Zhang, Shuai Jiang, Congzhentao Huang, Richard Yi Da Xu

In this paper, we propose a novel two-stream network with a lightweight resolution association ReID feature transformation (RAFT) module and a self-weighted attention (SWA) ReID module to evaluate features under different resolutions.

Person Re-Identification

Implicit bias of deep linear networks in the large learning rate phase

no code implementations25 Nov 2020 Wei Huang, Weitao Du, Richard Yi Da Xu, Chunrui Liu

We claim that depending on the separation conditions of data, the gradient descent iterates will converge to a flatter minimum in the catapult phase.

RGB-IR Cross-modality Person ReID based on Teacher-Student GAN Model

no code implementations15 Jul 2020 Ziyue Zhang, Shuai Jiang, Congzhentao Huang, Yang Li, Richard Yi Da Xu

To solve this challenge, we proposed a Teacher-Student GAN model (TS-GAN) to adopt different domains and guide the ReID backbone to learn better ReID information.

Person Re-Identification

On the Neural Tangent Kernel of Deep Networks with Orthogonal Initialization

3 code implementations13 Apr 2020 Wei Huang, Weitao Du, Richard Yi Da Xu

The prevailing thinking is that orthogonal weights are crucial to enforcing dynamical isometry and speeding up training.

Gaussian Process Latent Variable Model Factorization for Context-aware Recommender Systems

2 code implementations19 Dec 2019 Wei Huang, Richard Yi Da Xu

Our work is primarily inspired by the Gaussian Process Latent Variable Model (GPLVM), which was a non-linear dimensionality reduction method.

Dimensionality Reduction Recommendation Systems

Mean field theory for deep dropout networks: digging up gradient backpropagation deeply

1 code implementation19 Dec 2019 Wei Huang, Richard Yi Da Xu, Weitao Du, Yutian Zeng, Yunce Zhao

In recent years, the mean field theory has been applied to the study of neural networks and has achieved a great deal of success.

GAN-based Gaussian Mixture Model Responsibility Learning

no code implementations25 Sep 2019 Wanming Huang, Shuai Jiang, Xuan Liang, Ian Oppermann, Richard Yi Da Xu

Instead of defining p(x|k, θ) explicitly, we devised a modified GAN to allow us to define the distribution using p(z|k, θ), where z is the corresponding latent representation of x, as well as p(k|x, θ) through an additional classification network which is trained with the GAN in an “end-to-end” fashion.

Magnitude Bounded Matrix Factorisation for Recommender Systems

no code implementations15 Jul 2018 Shuai Jiang, Kan Li, Richard Yi Da Xu

Low rank matrix factorisation is often used in recommender systems as a way of extracting latent features.

Recommendation Systems

Diverse Online Feature Selection

1 code implementation12 Jun 2018 Chapman Siu, Richard Yi Da Xu

The framework aims to promote diversity based on the kernel produced on a feature level, through at most three stages: feature sampling, local criteria and global criteria for feature selection.

Point Processes

Cooperative Hierarchical Dirichlet Processes: Superposition vs. Maximization

no code implementations18 Jul 2017 Junyu Xuan, Jie Lu, Guangquan Zhang, Richard Yi Da Xu

The cooperative hierarchical structure is a common and significant data structure observed in, or adopted by, many research areas, such as: text mining (author-paper-word) and multi-label classification (label-instance-feature).

Multi-Label Classification Topic Models

The Dependent Random Measures with Independent Increments in Mixture Models

no code implementations27 Jun 2016 Cheng Luo, Richard Yi Da Xu, Yang Xiang

One of the propositions of the dependent random measures is that the atoms of the posterior distribution are shared amongst groups, and hence groups can borrow information from each other.

Smoothed Hierarchical Dirichlet Process: A Non-Parametric Approach to Constraint Measures

no code implementations16 Apr 2016 Cheng Luo, Yang Xiang, Richard Yi Da Xu

The key novelty of this model is that we place a temporal constraint amongst the nearby discrete measures $\{G_j\}$ in the form of symmetric Kullback-Leibler (KL) Divergence with a fixed bound $B$.

Dependent Indian Buffet Process-based Sparse Nonparametric Nonnegative Matrix Factorization

no code implementations12 Jul 2015 Junyu Xuan, Jie Lu, Guangquan Zhang, Richard Yi Da Xu, Xiangfeng Luo

Under this same framework, two classes of correlation function are proposed (1) using Bivariate beta distribution and (2) using Copula function.

Recommendation Systems

Infinite Author Topic Model based on Mixed Gamma-Negative Binomial Process

no code implementations30 Mar 2015 Junyu Xuan, Jie Lu, Guangquan Zhang, Richard Yi Da Xu, Xiangfeng Luo

One branch of these works is the so-called Author Topic Model (ATM), which incorporates the authors's interests as side information into the classical topic model.

Information Retrieval

An Adaptive Online HDP-HMM for Segmentation and Classification of Sequential Data

no code implementations10 Mar 2015 Ava Bargi, Richard Yi Da Xu, Massimo Piccardi

This infinite adaptive online approach is capable of segmenting and classifying the sequential data over unlimited number of classes, while meeting the memory and delay constraints of streaming contexts.

General Classification

Learning Hidden Structures with Relational Models by Adequately Involving Rich Information in A Network

no code implementations6 Oct 2013 Xuhui Fan, Richard Yi Da Xu, Longbing Cao, Yin Song

In this work, we propose an informative relational model (InfRM) framework to adequately involve rich information and its granularity in a network, including metadata information about each entity and various forms of link data.

A non-parametric conditional factor regression model for high-dimensional input and response

no code implementations2 Jul 2013 Ava Bargi, Richard Yi Da Xu, Massimo Piccardi

In this paper, we propose a non-parametric conditional factor regression (NCFR)model for domains with high-dimensional input and response.

Dimensionality Reduction

Dynamic Infinite Mixed-Membership Stochastic Blockmodel

no code implementations13 Jun 2013 Xuhui Fan, Longbing Cao, Richard Yi Da Xu

Directional and pairwise measurements are often used to model inter-relationships in a social network setting.

Copula Mixed-Membership Stochastic Blockmodel for Intra-Subgroup Correlations

no code implementations12 Jun 2013 Xuhui Fan, Longbing Cao, Richard Yi Da Xu

To this end, we introduce a \emph{Copula Mixed-Membership Stochastic Blockmodel (cMMSB)} where an individual Copula function is employed to jointly model the membership pairs of those nodes within the subgroup of interest.

Link Prediction

Cannot find the paper you are looking for? You can Submit a new open access paper.