Search Results for author: Bo Dai

Found 138 papers, 56 papers with code

3D Data Augmentation for Driving Scenes on Camera

no code implementations18 Mar 2023 Wenwen Tong, Jiangwei Xie, Tianyu Li, Hanming Deng, Xiangwei Geng, Ruoyi Zhou, Dingchen Yang, Bo Dai, Lewei Lu, Hongyang Li

The proposed data augmentation approach contributes to a gain of 1. 7% and 1. 4% in terms of detection accuracy, on Waymo and nuScences respectively.

Controllable Mesh Generation Through Sparse Latent Point Diffusion Models

no code implementations14 Mar 2023 Zhaoyang Lyu, Jinyi Wang, Yuwei An, Ya zhang, Dahua Lin, Bo Dai

In this work, we design a novel sparse latent point diffusion model for mesh generation.

Prototype-based Embedding Network for Scene Graph Generation

no code implementations13 Mar 2023 Chaofan Zheng, Xinyu Lyu, Lianli Gao, Bo Dai, Jingkuan Song

Current Scene Graph Generation (SGG) methods explore contextual information to predict relationships among entity pairs.

Graph Generation Scene Graph Generation

Learning Universal Policies via Text-Guided Video Generation

no code implementations31 Jan 2023 Yilun Du, Mengjiao Yang, Bo Dai, Hanjun Dai, Ofir Nachum, Joshua B. Tenenbaum, Dale Schuurmans, Pieter Abbeel

The proposed policy-as-video formulation can further represent environments with different state and action spaces in a unified space of images, which, for example, enables learning and generalization across a variety of robot manipulation tasks.

Decision Making Image Generation +3

Dynamic Storyboard Generation in an Engine-based Virtual Environment for Video Production

no code implementations30 Jan 2023 Anyi Rao, Xuekun Jiang, Yuwei Guo, Linning Xu, Lei Yang, Libiao Jin, Dahua Lin, Bo Dai

Amateurs working on mini-films and short-form videos usually spend lots of time and effort on the multi-round complicated process of setting and adjusting scenes, plots, and cameras to deliver satisfying video shots.

The Role of Baselines in Policy Gradient Optimization

no code implementations16 Jan 2023 Jincheng Mei, Wesley Chung, Valentin Thomas, Bo Dai, Csaba Szepesvari, Dale Schuurmans

Instead, the analysis reveals that the primary effect of the value baseline is to \textbf{reduce the aggressiveness of the updates} rather than their variance.

Correspondence Distillation from NeRF-based GAN

no code implementations19 Dec 2022 Yushi Lan, Chen Change Loy, Bo Dai

The neural radiance field (NeRF) has shown promising results in preserving the fine details of objects and scenes.

3DHumanGAN: Towards Photo-Realistic 3D-Aware Human Image Generation

1 code implementation14 Dec 2022 Zhuoqian Yang, Shikai Li, Wayne Wu, Bo Dai

We present 3DHumanGAN, a 3D-aware generative adversarial network (GAN) that synthesizes images of full-body humans with consistent appearances under different view-angles and body-poses.

Image Generation

Self-Supervised Geometry-Aware Encoder for Style-Based 3D GAN Inversion

no code implementations14 Dec 2022 Yushi Lan, Xuyi Meng, Shuai Yang, Chen Change Loy, Bo Dai

In this paper, we study the challenging problem of 3D GAN inversion where a latent code is predicted given a single face image to faithfully recover its 3D shapes and detailed textures.

3D Face Reconstruction

Score-based Continuous-time Discrete Diffusion Models

no code implementations30 Nov 2022 Haoran Sun, Lijun Yu, Bo Dai, Dale Schuurmans, Hanjun Dai

Score-based modeling through stochastic differential equations (SDEs) has provided a new perspective on diffusion models, and demonstrated superior performance on continuous data.

Oracle Inequalities for Model Selection in Offline Reinforcement Learning

no code implementations3 Nov 2022 Jonathan N. Lee, George Tucker, Ofir Nachum, Bo Dai, Emma Brunskill

We propose the first model selection algorithm for offline RL that achieves minimax rate-optimal oracle inequalities up to logarithmic factors.

Model Selection Offline RL +2

Factor Investing with a Deep Multi-Factor Model

no code implementations22 Oct 2022 Zikai Wei, Bo Dai, Dahua Lin

Modeling and characterizing multiple factors is perhaps the most important step in achieving excess returns over market benchmarks.

Graph Attention Management +1

Rethinking Trajectory Prediction via "Team Game"

no code implementations17 Oct 2022 Zikai Wei, Xinge Zhu, Bo Dai, Dahua Lin

To accurately predict trajectories in multi-agent settings, e. g. team games, it is important to effectively model the interactions among agents.

Trajectory Prediction

Temporal and Contextual Transformer for Multi-Camera Editing of TV Shows

no code implementations17 Oct 2022 Anyi Rao, Xuekun Jiang, Sichen Wang, Yuwei Guo, Zihao Liu, Bo Dai, Long Pang, Xiaoyu Wu, Dahua Lin, Libiao Jin

The ability to choose an appropriate camera view among multiple cameras plays a vital role in TV shows delivery.

Improving GANs with A Dynamic Discriminator

no code implementations20 Sep 2022 Ceyuan Yang, Yujun Shen, Yinghao Xu, Deli Zhao, Bo Dai, Bolei Zhou

Two capacity adjusting schemes are developed for training GANs under different data regimes: i) given a sufficient amount of training data, the discriminator benefits from a progressively increased learning capacity, and ii) when the training data is limited, gradually decreasing the layer width mitigates the over-fitting issue of the discriminator.

3D-Aware Image Synthesis Data Augmentation

Transformer with Implicit Edges for Particle-based Physics Simulation

1 code implementation22 Jul 2022 Yidi Shao, Chen Change Loy, Bo Dai

Consequently, in this paper we propose a novel Transformer-based method, dubbed as Transformer with Implicit Edges (TIE), to capture the rich semantics of particle interactions in an edge-free manner.

BRACE: The Breakdancing Competition Dataset for Dance Motion Synthesis

1 code implementation20 Jul 2022 Davide Moltisanti, Jinyi Wu, Bo Dai, Chen Change Loy

Estimating human keypoints from these videos is difficult due to the complexity of the dance, as well as the multiple moving cameras recording setup.

Motion Synthesis Pose Estimation

Monocular 3D Object Reconstruction with GAN Inversion

1 code implementation20 Jul 2022 Junzhe Zhang, Daxuan Ren, Zhongang Cai, Chai Kiat Yeo, Bo Dai, Chen Change Loy

Reconstruction is achieved by searching for a latent space in the 3D GAN that best resembles the target mesh in accordance with the single view observation.

3D Object Reconstruction

Making Linear MDPs Practical via Contrastive Representation Learning

no code implementations14 Jul 2022 Tianjun Zhang, Tongzheng Ren, Mengjiao Yang, Joseph E. Gonzalez, Dale Schuurmans, Bo Dai

It is common to address the curse of dimensionality in Markov decision processes (MDPs) by exploiting low-rank representations.

Representation Learning

Discrete Langevin Sampler via Wasserstein Gradient Flow

no code implementations29 Jun 2022 Haoran Sun, Hanjun Dai, Bo Dai, Haomin Zhou, Dale Schuurmans

It is known that gradient-based MCMC samplers for continuous spaces, such as Langevin Monte Carlo (LMC), can be derived as particle versions of a gradient flow that minimizes KL divergence on a Wasserstein manifold.

Guided Diffusion Model for Adversarial Purification

1 code implementation30 May 2022 Jinyi Wang, Zhaoyang Lyu, Dahua Lin, Bo Dai, Hongfei Fu

In this paper, we propose a novel purification approach, referred to as guided diffusion model for purification (GDMP), to help protect classifiers from adversarial attacks.


Towards Diverse and Natural Scene-aware 3D Human Motion Synthesis

no code implementations CVPR 2022 Jingbo Wang, Yu Rong, Jingyuan Liu, Sijie Yan, Dahua Lin, Bo Dai

The ability to synthesize long-term human motion sequences in real-world scenes can facilitate numerous applications.

Motion Synthesis

Accelerating Diffusion Models via Early Stop of the Diffusion Process

1 code implementation25 May 2022 Zhaoyang Lyu, Xudong Xu, Ceyuan Yang, Dahua Lin, Bo Dai

By modeling the reverse process of gradually diffusing the data distribution into a Gaussian distribution, generating a sample in DDPMs can be regarded as iteratively denoising a randomly sampled Gaussian noise.

Denoising Image Generation

DeciWatch: A Simple Baseline for 10x Efficient 2D and 3D Pose Estimation

1 code implementation16 Mar 2022 Ailing Zeng, Xuan Ju, Lei Yang, Ruiyuan Gao, Xizhou Zhu, Bo Dai, Qiang Xu

This paper proposes a simple baseline framework for video-based 2D/3D human pose estimation that can achieve 10 times efficiency improvement over existing works without any performance degradation, named DeciWatch.

2D Human Pose Estimation 3D Human Pose Estimation +2

SAFER: Data-Efficient and Safe Reinforcement Learning via Skill Acquisition

no code implementations10 Feb 2022 Dylan Slack, Yinlam Chow, Bo Dai, Nevan Wichers

However, we identify these techniques are not well equipped for safe policy learning because they ignore negative experiences(e. g., unsafe or unsuccessful), focusing only on positive experiences, which harms their ability to generalize to new tasks safely.

reinforcement-learning reinforcement Learning +2

Model Selection in Batch Policy Optimization

no code implementations23 Dec 2021 Jonathan N. Lee, George Tucker, Ofir Nachum, Bo Dai

We formalize the problem in the contextual bandit setting with linear model classes by identifying three sources of error that any model selection algorithm should optimally trade-off in order to be competitive: (1) approximation error, (2) statistical complexity, and (3) coverage.

Model Selection

Cross-Model Pseudo-Labeling for Semi-Supervised Action Recognition

no code implementations CVPR 2022 Yinghao Xu, Fangyun Wei, Xiao Sun, Ceyuan Yang, Yujun Shen, Bo Dai, Bolei Zhou, Stephen Lin

Typically in recent work, the pseudo-labels are obtained by training a model on the labeled data, and then using confident predictions from the model to teach itself.

Action Recognition

BungeeNeRF: Progressive Neural Radiance Field for Extreme Multi-scale Scene Rendering

no code implementations10 Dec 2021 Yuanbo Xiangli, Linning Xu, Xingang Pan, Nanxuan Zhao, Anyi Rao, Christian Theobalt, Bo Dai, Dahua Lin

The wide span of viewing positions within these scenes yields multi-scale renderings with very different levels of detail, which poses great challenges to neural radiance field and biases it towards compromised results.

Extract Free Dense Labels from CLIP

1 code implementation2 Dec 2021 Chong Zhou, Chen Change Loy, Bo Dai

Contrastive Language-Image Pre-training (CLIP) has made a remarkable breakthrough in open-vocabulary zero-shot image recognition.

Novel Concepts Semantic Segmentation +1

Towards understanding retrosynthesis by energy-based models

no code implementations NeurIPS 2021 Ruoxi Sun, Hanjun Dai, Li Li, Steven Kearnes, Bo Dai

In this paper, we propose a framework that unifies sequence- and graph-based methods as energy-based models (EBMs) with different energy functions.

Drug Discovery

Neural Stochastic Dual Dynamic Programming

no code implementations ICLR 2022 Hanjun Dai, Yuan Xue, Zia Syed, Dale Schuurmans, Bo Dai

Stochastic dual dynamic programming (SDDP) is a state-of-the-art method for solving multi-stage stochastic optimization, widely used for modeling real-world process optimization tasks.

Stochastic Optimization

A Free Lunch from the Noise: Provable and Practical Exploration for Representation Learning

no code implementations22 Nov 2021 Tongzheng Ren, Tianjun Zhang, Csaba Szepesvári, Bo Dai

Representation learning lies at the heart of the empirical success of deep learning for dealing with the curse of dimensionality.

Representation Learning

Deceive D: Adaptive Pseudo Augmentation for GAN Training with Limited Data

2 code implementations NeurIPS 2021 Liming Jiang, Bo Dai, Wayne Wu, Chen Change Loy

Generative adversarial networks (GANs) typically require ample data for training in order to synthesize high-fidelity images.

Generative Occupancy Fields for 3D Surface-Aware Image Synthesis

1 code implementation NeurIPS 2021 Xudong Xu, Xingang Pan, Dahua Lin, Bo Dai

In this paper, we propose Generative Occupancy Fields (GOF), a novel model based on generative radiance fields that can learn compact object surfaces without impeding its training convergence.

3D-Aware Image Synthesis

A Shading-Guided Generative Implicit Model for Shape-Accurate 3D-Aware Image Synthesis

1 code implementation NeurIPS 2021 Xingang Pan, Xudong Xu, Chen Change Loy, Christian Theobalt, Bo Dai

Motivated by the observation that a 3D object should look realistic from multiple viewpoints, these methods introduce a multi-view constraint as regularization to learn valid 3D radiance fields from 2D images.

3D-Aware Image Synthesis 3D Shape Reconstruction +1

Understanding the Effect of Stochasticity in Policy Optimization

no code implementations NeurIPS 2021 Jincheng Mei, Bo Dai, Chenjun Xiao, Csaba Szepesvari, Dale Schuurmans

We study the effect of stochasticity in on-policy policy optimization, and make the following four contributions.

SMORE: Knowledge Graph Completion and Multi-hop Reasoning in Massive Knowledge Graphs

1 code implementation28 Oct 2021 Hongyu Ren, Hanjun Dai, Bo Dai, Xinyun Chen, Denny Zhou, Jure Leskovec, Dale Schuurmans

There are two important reasoning tasks on KGs: (1) single-hop knowledge graph completion, which involves predicting individual links in the KG; and (2), multi-hop reasoning, where the goal is to predict which KG entities satisfy a given logical query.


MeshInversion: 3D textured mesh reconstruction with generative prior

no code implementations29 Sep 2021 Junzhe Zhang, Daxuan Ren, Zhongang Cai, Chai Kiat Yeo, Bo Dai, Chen Change Loy

Reconstruction is achieved by searching for a latent space in the 3D GAN that best resembles the target mesh in accordance with the single view observation.

Understanding and Leveraging Overparameterization in Recursive Value Estimation

no code implementations ICLR 2022 Chenjun Xiao, Bo Dai, Jincheng Mei, Oscar A Ramirez, Ramki Gummadi, Chris Harris, Dale Schuurmans

To better understand the utility of deep models in RL we present an analysis of recursive value estimation using overparameterized linear representations that provides useful, transferable findings.

Value prediction

SAFER: Data-Efficient and Safe Reinforcement Learning Through Skill Acquisition

no code implementations29 Sep 2021 Dylan Z Slack, Yinlam Chow, Bo Dai, Nevan Wichers

Though many reinforcement learning (RL) problems involve learning policies in settings that are difficult to specify safety constraints and sparse rewards, current methods struggle to rapidly and safely acquire successful policies.

reinforcement-learning reinforcement Learning +2

SiT: Simulation Transformer for Particle-based Physics Simulation

no code implementations29 Sep 2021 Yidi Shao, Chen Change Loy, Bo Dai

However, they force particles to interact with all neighbors without selection, and they fall short in capturing material semantics for different particles, leading to unsatisfactory performance, especially in generalization.

Combiner: Full Attention Transformer with Sparse Computation Cost

1 code implementation NeurIPS 2021 Hongyu Ren, Hanjun Dai, Zihang Dai, Mengjiao Yang, Jure Leskovec, Dale Schuurmans, Bo Dai

However, the key limitation of transformers is their quadratic memory and time complexity $\mathcal{O}(L^2)$ with respect to the sequence length in attention layers, which restricts application in extremely long sequences.

Image Generation Language Modelling

Safe Exploration by Solving Early Terminated MDP

no code implementations9 Jul 2021 Hao Sun, Ziping Xu, Meng Fang, Zhenghao Peng, Jiadong Guo, Bo Dai, Bolei Zhou

Safe exploration is crucial for the real-world application of reinforcement learning (RL).

Safe Exploration

The Curse of Passive Data Collection in Batch Reinforcement Learning

no code implementations18 Jun 2021 Chenjun Xiao, Ilbin Lee, Bo Dai, Dale Schuurmans, Csaba Szepesvari

In high stake applications, active experimentation may be considered too risky and thus data are often collected passively.

reinforcement-learning reinforcement Learning

Optimization Variance: Exploring Generalization Properties of DNNs

1 code implementation3 Jun 2021 Xiao Zhang, Dongrui Wu, Haoyi Xiong, Bo Dai

Unlike the conventional wisdom in statistical learning theory, the test error of a deep neural network (DNN) often demonstrates double descent: as the model complexity increases, it first follows a classical U-shaped curve and then shows a second descent.

Learning Theory

Scene-aware Generative Network for Human Motion Synthesis

no code implementations CVPR 2021 Jingbo Wang, Sijie Yan, Bo Dai, Dahua Lin

We revisit human motion synthesis, a task useful in various real world applications, in this paper.

Motion Synthesis

Leveraging Non-uniformity in First-order Non-convex Optimization

no code implementations13 May 2021 Jincheng Mei, Yue Gao, Bo Dai, Csaba Szepesvari, Dale Schuurmans

Classical global convergence results for first-order methods rely on uniform smoothness and the \L{}ojasiewicz inequality.

BIG-bench Machine Learning

Revisiting Skeleton-based Action Recognition

3 code implementations CVPR 2022 Haodong Duan, Yue Zhao, Kai Chen, Dahua Lin, Bo Dai

In this work, we propose PoseC3D, a new approach to skeleton-based action recognition, which relies on a 3D heatmap stack instead of a graph sequence as the base representation of human skeletons.

Action Recognition Group Activity Recognition +2

Unsupervised 3D Shape Completion through GAN Inversion

no code implementations CVPR 2021 Junzhe Zhang, Xinyi Chen, Zhongang Cai, Liang Pan, Haiyu Zhao, Shuai Yi, Chai Kiat Yeo, Bo Dai, Chen Change Loy

In contrast to previous fully supervised approaches, in this paper we present ShapeInversion, which introduces Generative Adversarial Network (GAN) inversion to shape completion for the first time.

Visually Informed Binaural Audio Generation without Binaural Audios

no code implementations CVPR 2021 Xudong Xu, Hang Zhou, Ziwei Liu, Bo Dai, Xiaogang Wang, Dahua Lin

Moreover, combined with binaural recordings, our method is able to further boost the performance of binaural audio generation under supervised settings.

Audio Generation

On the Optimality of Batch Policy Optimization Algorithms

no code implementations6 Apr 2021 Chenjun Xiao, Yifan Wu, Tor Lattimore, Bo Dai, Jincheng Mei, Lihong Li, Csaba Szepesvari, Dale Schuurmans

First, we introduce a class of confidence-adjusted index algorithms that unifies optimistic and pessimistic principles in a common framework, which enables a general analysis.

Value prediction

Nearly Horizon-Free Offline Reinforcement Learning

no code implementations NeurIPS 2021 Tongzheng Ren, Jialian Li, Bo Dai, Simon S. Du, Sujay Sanghavi

To the best of our knowledge, these are the \emph{first} set of nearly horizon-free bounds for episodic time-homogeneous offline tabular MDP and linear MDP with anchor points.

reinforcement-learning reinforcement Learning

DeepStyle: User Style Embedding for Authorship Attribution of Short Texts

no code implementations14 Mar 2021 Zhiqiang Hu, Roy Ka-Wei Lee, Lei Wang, Ee-Peng Lim, Bo Dai

Authorship attribution (AA), which is the task of finding the owner of a given text, is an important and widely studied research topic with many applications.

text-classification Text Classification

Off-Policy Imitation Learning from Observations

no code implementations NeurIPS 2020 Zhuangdi Zhu, Kaixiang Lin, Bo Dai, Jiayu Zhou

To further accelerate the learning procedure, we regulate the policy update with an inverse action model, which assists distribution matching from the perspective of mode-covering.

Imitation Learning

Towards Automatic Evaluation of Dialog Systems: A Model-Free Off-Policy Evaluation Approach

1 code implementation EMNLP 2021 Haoming Jiang, Bo Dai, Mengjiao Yang, Tuo Zhao, Wei Wei

An ideal environment for evaluating dialog systems, also known as the Turing test, needs to involve human interaction, which is usually not affordable for large-scale experiments.

Model-based Reinforcement Learning Off-policy evaluation +2

Self-Supervised Continuous Control without Policy Gradient

no code implementations1 Jan 2021 Hao Sun, Ziping Xu, Meng Fang, Yuhang Song, Jiechao Xiong, Bo Dai, Zhengyou Zhang, Bolei Zhou

Despite the remarkable progress made by the policy gradient algorithms in reinforcement learning (RL), sub-optimal policies usually result from the local exploration property of the policy gradient update.

Continuous Control Policy Gradient Methods +3

Slow Control System for PandaX-III experiment

no code implementations24 Dec 2020 Xiyu Yan, Xun Chen, Yu Chen, Bo Dai, Heng Lin, Tao Li, Ke Han, Kaixiang Ni, Fusang Wang, Shaobo Wang, Qibin Zheng, Xinning Zeng

The PandaX-III experiment uses high pressure gaseous time projection chamber to search for the neutrinoless double beta decay of $^{136}$Xe.

Anomaly Detection High Energy Physics - Experiment Instrumentation and Detectors

Offline Policy Selection under Uncertainty

1 code implementation12 Dec 2020 Mengjiao Yang, Bo Dai, Ofir Nachum, George Tucker, Dale Schuurmans

More importantly, we show how the belief distribution estimated by BayesDICE may be used to rank policies with respect to any arbitrary downstream policy selection metric, and we empirically demonstrate that this selection procedure significantly outperforms existing approaches, such as ranking policies according to mean or high-confidence lower bound value estimates.

Escaping the Gravitational Pull of Softmax

no code implementations NeurIPS 2020 Jincheng Mei, Chenjun Xiao, Bo Dai, Lihong Li, Csaba Szepesvari, Dale Schuurmans

Both findings are based on an analysis of convergence rates using the Non-uniform \L{}ojasiewicz (N\L{}) inequalities.

Differentiable Top-k with Optimal Transport

no code implementations NeurIPS 2020 Yujia Xie, Hanjun Dai, Minshuo Chen, Bo Dai, Tuo Zhao, Hongyuan Zha, Wei Wei, Tomas Pfister

Finding the k largest or smallest elements from a collection of scores, i. e., top-k operation, is an important model component widely used in information retrieval, machine learning, and data mining.

Information Retrieval Retrieval

Provably Efficient Neural Estimation of Structural Equation Models: An Adversarial Approach

no code implementations NeurIPS 2020 Luofeng Liao, You-Lin Chen, Zhuoran Yang, Bo Dai, Mladen Kolar, Zhaoran Wang

We study estimation in a class of generalized SEMs where the object of interest is defined as the solution to a linear operator equation.

Learning Discrete Energy-based Models via Auxiliary-variable Local Exploration

no code implementations NeurIPS 2020 Hanjun Dai, Rishabh Singh, Bo Dai, Charles Sutton, Dale Schuurmans

In this paper we propose ALOE, a new algorithm for learning conditional and unconditional EBMs for discrete structured data, where parameter gradients are estimated using a learned sampler that mimics local search.

Language Modelling

Do 2D GANs Know 3D Shape? Unsupervised 3D shape reconstruction from 2D Image GANs

1 code implementation ICLR 2021 Xingang Pan, Bo Dai, Ziwei Liu, Chen Change Loy, Ping Luo

Through our investigation, we found that such a pre-trained GAN indeed contains rich 3D knowledge and thus can be used to recover 3D shape from a single 2D image in an unsupervised manner.

3D Shape Reconstruction

Named Entity Recognition for Social Media Texts with Semantic Augmentation

1 code implementation EMNLP 2020 Yuyang Nie, Yuanhe Tian, Xiang Wan, Yan Song, Bo Dai

In particular, we obtain the augmented semantic information from a large-scale corpus, and propose an attentive semantic augmentation module and a gate module to encode and aggregate such information, respectively.

Chinese Named Entity Recognition named-entity-recognition +3

CoinDICE: Off-Policy Confidence Interval Estimation

no code implementations NeurIPS 2020 Bo Dai, Ofir Nachum, Yinlam Chow, Lihong Li, Csaba Szepesvári, Dale Schuurmans

We study high-confidence behavior-agnostic off-policy evaluation in reinforcement learning, where the goal is to estimate a confidence interval on a target policy's value, given only access to a static experience dataset collected by unknown behavior policies.

Off-policy evaluation

Differentiable Top-$k$ with Optimal Transport

no code implementations NeurIPS Workshop LMCA 2020 Yujia Xie, Hanjun Dai, Minshuo Chen, Bo Dai, Tuo Zhao, Hongyuan Zha, Wei Wei, Tomas Pfister

The top-$k$ operation, i. e., finding the $k$ largest or smallest elements from a collection of scores, is an important model component, which is widely used in information retrieval, machine learning, and data mining.

Information Retrieval Retrieval

Small Towers Make Big Differences

no code implementations13 Aug 2020 Yuyan Wang, Zhe Zhao, Bo Dai, Christopher Fifty, Dong Lin, Lichan Hong, Ed H. Chi

A delicate balance between multi-task generalization and multi-objective optimization is therefore needed for finding a better trade-off between efficiency and generalization.

Multi-Task Learning

Energy-based View of Retrosynthesis

no code implementations14 Jul 2020 Ruoxi Sun, Hanjun Dai, Li Li, Steven Kearnes, Bo Dai

Retrosynthesis -- the process of identifying a set of reactants to synthesize a target molecule -- is of vital importance to material design and drug discovery.

Drug Discovery Single-step retrosynthesis

Off-Policy Evaluation via the Regularized Lagrangian

no code implementations NeurIPS 2020 Mengjiao Yang, Ofir Nachum, Bo Dai, Lihong Li, Dale Schuurmans

The recently proposed distribution correction estimation (DICE) family of estimators has advanced the state of the art in off-policy evaluation from behavior-agnostic data.

Off-policy evaluation

Provably Efficient Neural Estimation of Structural Equation Model: An Adversarial Approach

no code implementations2 Jul 2020 Luofeng Liao, You-Lin Chen, Zhuoran Yang, Bo Dai, Zhaoran Wang, Mladen Kolar

We study estimation in a class of generalized SEMs where the object of interest is defined as the solution to a linear operator equation.

Unsupervised Landmark Learning from Unpaired Data

1 code implementation29 Jun 2020 Yinghao Xu, Ceyuan Yang, Ziwei Liu, Bo Dai, Bolei Zhou

Recent attempts for unsupervised landmark learning leverage synthesized image pairs that are similar in appearance but different in poses.

Scalable Deep Generative Modeling for Sparse Graphs

1 code implementation ICML 2020 Hanjun Dai, Azade Nazi, Yujia Li, Bo Dai, Dale Schuurmans

Based on this, we develop a novel autoregressive model, named BiGG, that utilizes this sparsity to avoid generating the full adjacency matrix, and importantly reduces the graph generation time complexity to $O((n + m)\log n)$.

Graph Generation

Video Representation Learning with Visual Tempo Consistency

1 code implementation28 Jun 2020 Ceyuan Yang, Yinghao Xu, Bo Dai, Bolei Zhou

Visual tempo, which describes how fast an action goes, has shown its potential in supervised action recognition.

Action Anticipation Action Detection +3

Zeroth-Order Supervised Policy Improvement

no code implementations11 Jun 2020 Hao Sun, Ziping Xu, Yuhang Song, Meng Fang, Jiechao Xiong, Bo Dai, Bolei Zhou

However, PG algorithms rely on exploiting the value function being learned with the first-order update locally, which results in limited sample efficiency.

Continuous Control Policy Gradient Methods +1

Novel Policy Seeking with Constrained Optimization

1 code implementation21 May 2020 Hao Sun, Zhenghao Peng, Bo Dai, Jian Guo, Dahua Lin, Bolei Zhou

In problem-solving, we humans can come up with multiple novel solutions to the same problem.

reinforcement-learning reinforcement Learning

Intra- and Inter-Action Understanding via Temporal Action Parsing

no code implementations CVPR 2020 Dian Shao, Yue Zhao, Bo Dai, Dahua Lin

Current methods for action recognition primarily rely on deep convolutional networks to derive feature embeddings of visual and motion features.

Action Parsing Action Recognition +1

Evolutionary Stochastic Policy Distillation

2 code implementations27 Apr 2020 Hao Sun, Xinyu Pan, Bo Dai, Dahua Lin, Bolei Zhou

Solving the Goal-Conditioned Reward Sparse (GCRS) task is a challenging reinforcement learning problem due to the sparsity of reward signals.

Temporal Pyramid Network for Action Recognition

3 code implementations CVPR 2020 Ceyuan Yang, Yinghao Xu, Jianping Shi, Bo Dai, Bolei Zhou

Previous works often capture the visual tempo through sampling raw videos at multiple rates and constructing an input-level frame pyramid, which usually requires a costly multi-branch network to handle.

Action Recognition

Self-Supervised Scene De-occlusion

2 code implementations CVPR 2020 Xiaohang Zhan, Xingang Pan, Bo Dai, Ziwei Liu, Dahua Lin, Chen Change Loy

This is achieved via Partial Completion Network (PCNet)-mask (M) and -content (C), that learn to recover fractions of object masks and contents, respectively, in a self-supervised manner.

Image Manipulation Scene Understanding

Learning Sparse Rewarded Tasks from Sub-Optimal Demonstrations

1 code implementation1 Apr 2020 Zhuangdi Zhu, Kaixiang Lin, Bo Dai, Jiayu Zhou

SAIL bridges the advantages of IL and RL to reduce the sample complexity substantially, by effectively exploiting sup-optimal demonstrations and efficiently exploring the environment to surpass the demonstrated performance.

Continuous Control Imitation Learning

Energy-Based Processes for Exchangeable Data

1 code implementation ICML 2020 Mengjiao Yang, Bo Dai, Hanjun Dai, Dale Schuurmans

Recently there has been growing interest in modeling sets with exchangeability such as point clouds.

Denoising Point Cloud Generation

Batch Stationary Distribution Estimation

1 code implementation ICML 2020 Junfeng Wen, Bo Dai, Lihong Li, Dale Schuurmans

We consider the problem of approximating the stationary distribution of an ergodic Markov chain given a set of sampled transitions.

Off-policy evaluation

GenDICE: Generalized Offline Estimation of Stationary Values

1 code implementation ICLR 2020 Ruiyi Zhang, Bo Dai, Lihong Li, Dale Schuurmans

An important problem that arises in reinforcement learning and Monte Carlo methods is estimating quantities defined by the stationary distribution of a Markov chain.

Differentiable Top-k Operator with Optimal Transport

no code implementations16 Feb 2020 Yujia Xie, Hanjun Dai, Minshuo Chen, Bo Dai, Tuo Zhao, Hongyuan Zha, Wei Wei, Tomas Pfister

The top-k operation, i. e., finding the k largest or smallest elements from a collection of scores, is an important model component, which is widely used in information retrieval, machine learning, and data mining.

Information Retrieval Retrieval

Real or Not Real, that is the Question

2 code implementations ICLR 2020 Yuanbo Xiangli, Yubin Deng, Bo Dai, Chen Change Loy, Dahua Lin

While generative adversarial networks (GAN) have been widely adopted in various topics, in this paper we generalize the standard GAN to a new perspective by treating realness as a random variable that can be estimated from multiple angles.

Reinforcement Learning via Fenchel-Rockafellar Duality

1 code implementation7 Jan 2020 Ofir Nachum, Bo Dai

We review basic concepts of convex duality, focusing on the very general and supremely useful Fenchel-Rockafellar duality.

reinforcement-learning reinforcement Learning

AlgaeDICE: Policy Gradient from Arbitrary Experience

no code implementations4 Dec 2019 Ofir Nachum, Bo Dai, Ilya Kostrikov, Yin-Lam Chow, Lihong Li, Dale Schuurmans

In many real-world applications of reinforcement learning (RL), interactions with the environment are limited due to cost or feasibility.

Overcoming Catastrophic Forgetting by Generative Regularization

no code implementations3 Dec 2019 Patrick H. Chen, Wei Wei, Cho-Jui Hsieh, Bo Dai

In this paper, we propose a new method to overcome catastrophic forgetting by adding generative regularization to Bayesian inference framework.

Bayesian Inference Continual Learning

Energy-Inspired Models: Learning with Sampler-Induced Distributions

1 code implementation NeurIPS 2019 Dieterich Lawson, George Tucker, Bo Dai, Rajesh Ranganath

Motivated by this, we consider the sampler-induced distribution as the model of interest and maximize the likelihood of this model.

Variational Inference

Learning with Social Influence through Interior Policy Differentiation

no code implementations25 Sep 2019 Hao Sun, Bo Dai, Jiankai Sun, Zhenghao Peng, Guodong Xu, Dahua Lin, Bolei Zhou

In this work we model the social influence into the scheme of reinforcement learning, enabling the agents to learn both from the environment and from their peers.

reinforcement Learning

Recursive Visual Sound Separation Using Minus-Plus Net

no code implementations ICCV 2019 Xudong Xu, Bo Dai, Dahua Lin

Sounds provide rich semantics, complementary to visual data, for many tasks.

DualDICE: Behavior-Agnostic Estimation of Discounted Stationary Distribution Corrections

2 code implementations NeurIPS 2019 Ofir Nachum, Yin-Lam Chow, Bo Dai, Lihong Li

In contrast to previous approaches, our algorithm is agnostic to knowledge of the behavior policy (or policies) used to generate the dataset.

Cooperative neural networks (CoNN): Exploiting prior independence structure for improved classification

no code implementations NeurIPS 2018 Harsh Shrivastava, Eugene Bart, Bob Price, Hanjun Dai, Bo Dai, Srinivas Aluru

We propose a new approach, called cooperative neural networks (CoNN), which uses a set of cooperatively trained neural networks to capture latent representations that exploit prior given independence structure.

General Classification text-classification +1

Exponential Family Estimation via Adversarial Dynamics Embedding

1 code implementation NeurIPS 2019 Bo Dai, Zhen Liu, Hanjun Dai, Niao He, Arthur Gretton, Le Song, Dale Schuurmans

We present an efficient algorithm for maximum likelihood estimation (MLE) of exponential family models, with a general parametrization of the energy function that includes neural networks.

Feature Intertwiner for Object Detection

2 code implementations ICLR 2019 Hongyang Li, Bo Dai, Shaoshuai Shi, Wanli Ouyang, Xiaogang Wang

We argue that the reliable set could guide the feature learning of the less reliable set during training - in spirit of student mimicking teacher behavior and thus pushing towards a more compact class centroid in the feature space.

object-detection Object Detection

Revisiting Auxiliary Latent Variables in Generative Models

no code implementations ICLR Workshop DeepGenStruct 2019 Dieterich Lawson, George Tucker, Bo Dai, Rajesh Ranganath

The success of enriching the variational family with auxiliary latent variables motivates applying the same techniques to the generative model.

Learning to Defense by Learning to Attack

no code implementations ICLR Workshop DeepGenStruct 2019 Zhehui Chen, Haoming Jiang, Yuyang Shi, Bo Dai, Tuo Zhao

From the perspective of generative learning, our proposed method can be viewed as learning a deep generative model for generating adversarial samples, which is adaptive to the robust classification.

Adversarial Attack Robust classification

Learning to Plan in High Dimensions via Neural Exploration-Exploitation Trees

1 code implementation ICLR 2020 Binghong Chen, Bo Dai, Qinjie Lin, Guo Ye, Han Liu, Le Song

We propose a meta path planning algorithm named \emph{Neural Exploration-Exploitation Trees~(NEXT)} for learning from prior experience for solving new path planning problems in high dimensional continuous state and action spaces.

Meta Architecture Search

1 code implementation NeurIPS 2019 Albert Shaw, Wei Wei, Weiyang Liu, Le Song, Bo Dai

Neural Architecture Search (NAS) has been quite successful in constructing state-of-the-art models on a variety of tasks.

Bayesian Inference Few-Shot Learning +1

Coupled Variational Bayes via Optimization Embedding

1 code implementation NeurIPS 2018 Bo Dai, Hanjun Dai, Niao He, Weiyang Liu, Zhen Liu, Jianshu Chen, Lin Xiao, Le Song

This flexible function class couples the variational distribution with the original parameters in the graphical models, allowing end-to-end learning of the graphical models by back-propagation through the variational distribution.

Variational Inference

Predictive Approximate Bayesian Computation via Saddle Points

no code implementations NeurIPS 2018 Yingxiang Yang, Bo Dai, Negar Kiyavash, Niao He

Approximate Bayesian computation (ABC) is an important methodology for Bayesian inference when the likelihood function is intractable.

Bayesian Inference regression

Kernel Exponential Family Estimation via Doubly Dual Embedding

1 code implementation6 Nov 2018 Bo Dai, Hanjun Dai, Arthur Gretton, Le Song, Dale Schuurmans, Niao He

We investigate penalized maximum log-likelihood estimation for exponential family distributions whose natural parameter resides in a reproducing kernel Hilbert space.

Learning to Defend by Learning to Attack

no code implementations3 Nov 2018 Haoming Jiang, Zhehui Chen, Yuyang Shi, Bo Dai, Tuo Zhao

Adversarial training provides a principled approach for training robust neural networks.

Adversarial Attack Adversarial Defense +2

A Neural Compositional Paradigm for Image Captioning

1 code implementation NeurIPS 2018 Bo Dai, Sanja Fidler, Dahua Lin

Mainstream captioning models often follow a sequential structure to generate captions, leading to issues such as introduction of irrelevant semantics, lack of diversity in the generated captions, and inadequate generalization performance.

Image Captioning

Neural Network Encapsulation

2 code implementations ECCV 2018 Hongyang Li, Xiaoyang Guo, Bo Dai, Wanli Ouyang, Xiaogang Wang

Motivated by the routing to make higher capsule have agreement with lower capsule, we extend the mechanism as a compensation for the rapid loss of information in nearby layers.

Move Forward and Tell: A Progressive Generator of Video Descriptions

no code implementations ECCV 2018 Yilei Xiong, Bo Dai, Dahua Lin

We present an efficient framework that can generate a coherent paragraph to describe a given video.

Video Captioning

Rethinking the Form of Latent States in Image Captioning

no code implementations ECCV 2018 Bo Dai, Deming Ye, Dahua Lin

Taking advantage of this, we visually reveal the internal dynamics in the process of caption generation, as well as the connections between input visual domain and output linguistic domain.

Image Captioning

Learning Deep Hidden Nonlinear Dynamics from Aggregate Data

no code implementations22 Jul 2018 Yisen Wang, Bo Dai, Lingkai Kong, Sarah Monazam Erfani, James Bailey, Hongyuan Zha

Learning nonlinear dynamics from diffusion data is a challenging problem since the individuals observed may be different at different time points, generally following an aggregate behaviour.

Learning Steady-States of Iterative Algorithms over Graphs

no code implementations ICML 2018 Hanjun Dai, Zornitsa Kozareva, Bo Dai, Alex Smola, Le Song

Many graph analytics problems can be solved via iterative algorithms where the solutions are often characterized by a set of steady-state conditions.

Learning towards Minimum Hyperspherical Energy

4 code implementations NeurIPS 2018 Weiyang Liu, Rongmei Lin, Zhen Liu, Lixin Liu, Zhiding Yu, Bo Dai, Le Song

In light of this intuition, we reduce the redundancy regularization problem to generic energy minimization, and propose a minimum hyperspherical energy (MHE) objective as generic regularization for neural networks.

Decoupled Networks

1 code implementation CVPR 2018 Weiyang Liu, Zhen Liu, Zhiding Yu, Bo Dai, Rongmei Lin, Yisen Wang, James M. Rehg, Le Song

Inner product-based convolution has been a central component of convolutional neural networks (CNNs) and the key to learning visual representations.

SBEED: Convergent Reinforcement Learning with Nonlinear Function Approximation

no code implementations ICML 2018 Bo Dai, Albert Shaw, Lihong Li, Lin Xiao, Niao He, Zhen Liu, Jianshu Chen, Le Song

When function approximation is used, solving the Bellman optimality equation with stability guarantees has remained a major open problem in reinforcement learning for decades.

Q-Learning reinforcement-learning +1

Boosting the Actor with Dual Critic

no code implementations ICLR 2018 Bo Dai, Albert Shaw, Niao He, Lihong Li, Le Song

This paper proposes a new actor-critic-style algorithm called Dual Actor-Critic or Dual-AC.

Deep Hyperspherical Learning

no code implementations NeurIPS 2017 Weiyang Liu, Yan-Ming Zhang, Xingguo Li, Zhiding Yu, Bo Dai, Tuo Zhao, Le Song

In light of such challenges, we propose hyperspherical convolution (SphereConv), a novel learning framework that gives angular representations on hyperspheres.

Representation Learning

Towards Black-box Iterative Machine Teaching

no code implementations ICML 2018 Weiyang Liu, Bo Dai, Xingguo Li, Zhen Liu, James M. Rehg, Le Song

We propose an active teacher model that can actively query the learner (i. e., make the learner take exams) for estimating the learner's status and provably guide the learner to achieve faster convergence.

Contrastive Learning for Image Captioning

no code implementations NeurIPS 2017 Bo Dai, Dahua Lin

Specifically, via two constraints formulated on top of a reference model, the proposed method can encourage distinctiveness, while maintaining the overall quality of the generated captions.

Contrastive Learning Image Captioning

Iterative Machine Teaching

2 code implementations ICML 2017 Weiyang Liu, Bo Dai, Ahmad Humayun, Charlene Tay, Chen Yu, Linda B. Smith, James M. Rehg, Le Song

Different from traditional machine teaching which views the learners as batch algorithms, we study a new paradigm where the learner uses an iterative algorithm and a teacher can feed examples sequentially and intelligently based on the current performance of the learner.

Towards Diverse and Natural Image Descriptions via a Conditional GAN

1 code implementation ICCV 2017 Bo Dai, Sanja Fidler, Raquel Urtasun, Dahua Lin

Despite the substantial progress in recent years, the image captioning techniques are still far from being perfect. Sentences produced by existing methods, e. g. those based on RNNs, are often overly rigid and lacking in variability.

Image Captioning

Stochastic Generative Hashing

2 code implementations ICML 2017 Bo Dai, Ruiqi Guo, Sanjiv Kumar, Niao He, Le Song

Learning-based binary hashing has become a powerful paradigm for fast search and retrieval in massive databases.


Learning from Conditional Distributions via Dual Embeddings

no code implementations15 Jul 2016 Bo Dai, Niao He, Yunpeng Pan, Byron Boots, Le Song

In such problems, each sample $x$ itself is associated with a conditional distribution $p(z|x)$ represented by samples $\{z_i\}_{i=1}^M$, and the goal is to learn a function $f$ that links these conditional distributions to target values $y$.

Discriminative Embeddings of Latent Variable Models for Structured Data

1 code implementation17 Mar 2016 Hanjun Dai, Bo Dai, Le Song

Kernel classifiers and regressors designed for structured data, such as sequences, trees and graphs, have significantly advanced a number of interdisciplinary areas such as computational biology and drug design.

Provable Bayesian Inference via Particle Mirror Descent

no code implementations9 Jun 2015 Bo Dai, Niao He, Hanjun Dai, Le Song

Bayesian methods are appealing in their flexibility in modeling complex data and ability in capturing uncertainty in parameters.

Bayesian Inference Gaussian Processes

Scalable Kernel Methods via Doubly Stochastic Gradients

1 code implementation NeurIPS 2014 Bo Dai, Bo Xie, Niao He, YIngyu Liang, Anant Raj, Maria-Florina Balcan, Le Song

The general perception is that kernel methods are not scalable, and neural nets are the methods of choice for nonlinear learning problems.

Transductive Learning with Multi-class Volume Approximation

no code implementations3 Feb 2014 Gang Niu, Bo Dai, Marthinus Christoffel du Plessis, Masashi Sugiyama

Given a hypothesis space, the large volume principle by Vladimir Vapnik prioritizes equivalence classes according to their volume in the hypothesis space.

Transductive Learning

Nonparametric Estimation of Multi-View Latent Variable Models

no code implementations13 Nov 2013 Le Song, Animashree Anandkumar, Bo Dai, Bo Xie

We establish that the sample complexity for the proposed method is quadratic in the number of latent components and is a low order polynomial in the other relevant parameters.

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