no code implementations • EMNLP 2020 • Zhengjue Wang, Zhibin Duan, Hao Zhang, Chaojie Wang, Long Tian, Bo Chen, Mingyuan Zhou
Abstractive document summarization is a comprehensive task including document understanding and summary generation, in which area Transformer-based models have achieved the state-of-the-art performance.
no code implementations • 7 May 2024 • Shujian Zhang, Korawat Tanwisuth, Chengyue Gong, Pengcheng He, Mingyuan Zhou
Auto-regressive generation models achieve competitive performance across many different NLP tasks such as summarization, question answering, and classifications.
no code implementations • 5 Apr 2024 • Mingyuan Zhou, Huangjie Zheng, Zhendong Wang, Mingzhang Yin, Hai Huang
This achievement not only redefines the benchmarks for efficiency and effectiveness in diffusion distillation but also in the broader field of diffusion-based generation.
1 code implementation • 22 Feb 2024 • Xuxi Chen, Zhendong Wang, Daouda Sow, Junjie Yang, Tianlong Chen, Yingbin Liang, Mingyuan Zhou, Zhangyang Wang
Our study starts from an empirical strategy for the light continual training of LLMs using their original pre-training data sets, with a specific focus on selective retention of samples that incur moderately high losses.
1 code implementation • 13 Feb 2024 • Shentao Yang, Tianqi Chen, Mingyuan Zhou
Aligning text-to-image diffusion model (T2I) with preference has been gaining increasing research attention.
1 code implementation • 12 Feb 2024 • Yueqin Yin, Zhendong Wang, Yi Gu, Hai Huang, Weizhu Chen, Mingyuan Zhou
In the field of large language models (LLMs), aligning models with the diverse preferences of users is a critical challenge.
no code implementations • 20 Jan 2024 • Mingyuan Zhou, Rakib Hyder, Ziwei Xuan, GuoJun Qi
Recent advances in 3D avatar generation have gained significant attentions.
no code implementations • 3 Dec 2023 • Tianqi Chen, Yongfei Liu, Zhendong Wang, Jianbo Yuan, Quanzeng You, Hongxia Yang, Mingyuan Zhou
In light of the remarkable success of in-context learning in large language models, its potential extension to the vision domain, particularly with visual foundation models like Stable Diffusion, has sparked considerable interest.
no code implementations • 30 Nov 2023 • Zhangsihao Yang, Mingyuan Zhou, Mengyi Shan, Bingbing Wen, Ziwei Xuan, Mitch Hill, Junjie Bai, Guo-Jun Qi, Yalin Wang
Our paper aims to generate diverse and realistic animal motion sequences from textual descriptions, without a large-scale animal text-motion dataset.
no code implementations • 10 Oct 2023 • Huangjie Zheng, Zhendong Wang, Jianbo Yuan, Guanghan Ning, Pengcheng He, Quanzeng You, Hongxia Yang, Mingyuan Zhou
Diffusion models excel at generating photo-realistic images but come with significant computational costs in both training and sampling.
1 code implementation • NeurIPS 2023 • Mingyuan Zhou, Tianqi Chen, Zhendong Wang, Huangjie Zheng
We introduce beta diffusion, a novel generative modeling method that integrates demasking and denoising to generate data within bounded ranges.
1 code implementation • ICCV 2023 • Miaoge Li, Dongsheng Wang, Xinyang Liu, Zequn Zeng, Ruiying Lu, Bo Chen, Mingyuan Zhou
We find that by formulating the multi-label classification as a CT problem, we can exploit the interactions between the image and label efficiently by minimizing the bidirectional CT cost.
1 code implementation • 28 May 2023 • Tianqi Chen, Mingyuan Zhou
However, it is found in this paper to have limited ability in modeling some other types of data, such as count and non-negative continuous data, that are often highly sparse, skewed, heavy-tailed, and/or overdispersed.
no code implementations • 4 May 2023 • Shujian Zhang, Chengyue Gong, Lemeng Wu, Xingchao Liu, Mingyuan Zhou
Ultimately, with this prompt paragraph, AutoML-GPT will automatically conduct the experiments from data processing to model architecture, hyperparameter tuning, and predicted training log.
1 code implementation • NeurIPS 2023 • Zhendong Wang, Yifan Jiang, Yadong Lu, Yelong Shen, Pengcheng He, Weizhu Chen, Zhangyang Wang, Mingyuan Zhou
We present Prompt Diffusion, a framework for enabling in-context learning in diffusion-based generative models.
1 code implementation • CVPR 2023 • Yiming Qin, Huangjie Zheng, Jiangchao Yao, Mingyuan Zhou, Ya zhang
To tackle this problem, we set from the hypothesis that the data distribution is not class-balanced, and propose Class-Balancing Diffusion Models (CBDM) that are trained with a distribution adjustment regularizer as a solution.
1 code implementation • 29 Apr 2023 • Korawat Tanwisuth, Shujian Zhang, Huangjie Zheng, Pengcheng He, Mingyuan Zhou
Through prompting, large-scale pre-trained models have become more expressive and powerful, gaining significant attention in recent years.
1 code implementation • NeurIPS 2023 • Zhendong Wang, Yifan Jiang, Huangjie Zheng, Peihao Wang, Pengcheng He, Zhangyang Wang, Weizhu Chen, Mingyuan Zhou
Patch Diffusion meanwhile improves the performance of diffusion models trained on relatively small datasets, $e. g.$, as few as 5, 000 images to train from scratch.
1 code implementation • 11 Apr 2023 • Mohammadreza Armandpour, Ali Sadeghian, Huangjie Zheng, Amir Sadeghian, Mingyuan Zhou
Although text-to-image diffusion models have made significant strides in generating images from text, they are sometimes more inclined to generate images like the data on which the model was trained rather than the provided text.
no code implementations • 16 Mar 2023 • Xinyang Liu, Dongsheng Wang, Miaoge Li, Zhibin Duan, Yishi Xu, Bo Chen, Mingyuan Zhou
For downstream applications of vision-language pre-trained models, there has been significant interest in constructing effective prompts.
1 code implementation • CVPR 2023 • Zhixin Wang, Xiaoyun Zhang, Ziying Zhang, Huangjie Zheng, Mingyuan Zhou, Ya zhang, Yanfeng Wang
However, it is expensive and infeasible to include every type of degradation to cover real-world cases in the training data.
2 code implementations • 20 Feb 2023 • Yihao Feng, Shentao Yang, Shujian Zhang, JianGuo Zhang, Caiming Xiong, Mingyuan Zhou, Huan Wang
Prior works mainly focus on adopting advanced RL techniques to train the ToD agents, while the design of the reward function is not well studied.
no code implementations • 8 Feb 2023 • Korawat Tanwisuth, Shujian Zhang, Pengcheng He, Mingyuan Zhou
Finally, it refines the target model on the target domain data without guidance from the source model.
no code implementations • 11 Jan 2023 • Chengzhi Wu, Julius Pfrommer, Mingyuan Zhou, Jürgen Beyerer
We propose a combined generative and contrastive neural architecture for learning latent representations of 3D volumetric shapes.
1 code implementation • 16 Oct 2022 • Yishi Xu, Dongsheng Wang, Bo Chen, Ruiying Lu, Zhibin Duan, Mingyuan Zhou
With the tree-likeness property of hyperbolic space, the underlying semantic hierarchy among words and topics can be better exploited to mine more interpretable topics.
1 code implementation • 12 Oct 2022 • Shentao Yang, Shujian Zhang, Yihao Feng, Mingyuan Zhou
In offline model-based reinforcement learning (offline MBRL), we learn a dynamic model from historically collected data, and subsequently utilize the learned model and fixed datasets for policy learning, without further interacting with the environment.
no code implementations • 9 Oct 2022 • Dandan Guo, Long Tian, He Zhao, Mingyuan Zhou, Hongyuan Zha
A recent solution to this problem is calibrating the distribution of these few sample classes by transferring statistics from the base classes with sufficient examples, where how to decide the transfer weights from base classes to novel classes is the key.
1 code implementation • 20 Sep 2022 • Dongsheng Wang, Yishi Xu, Miaoge Li, Zhibin Duan, Chaojie Wang, Bo Chen, Mingyuan Zhou
We propose a Bayesian generative model for incorporating prior domain knowledge into hierarchical topic modeling.
no code implementations • 12 Sep 2022 • Dongsheng Wang, Chaojie Wang, Bo Chen, Mingyuan Zhou
To build recommender systems that not only consider user-item interactions represented as ordinal variables, but also exploit the social network describing the relationships between the users, we develop a hierarchical Bayesian model termed ordinal graph factor analysis (OGFA), which jointly models user-item and user-user interactions.
3 code implementations • 12 Aug 2022 • Zhendong Wang, Jonathan J Hunt, Mingyuan Zhou
In our approach, we learn an action-value function and we add a term maximizing action-values into the training loss of the conditional diffusion model, which results in a loss that seeks optimal actions that are near the behavior policy.
no code implementations • 5 Aug 2022 • Dandan Guo, Zhuo Li, Meixi Zheng, He Zhao, Mingyuan Zhou, Hongyuan Zha
Specifically, we view the training set as an imbalanced distribution over its samples, which is transported by OT to a balanced distribution obtained from the meta set.
2 code implementations • 15 Jun 2022 • Xizewen Han, Huangjie Zheng, Mingyuan Zhou
In this paper, we introduce classification and regression diffusion (CARD) models, which combine a denoising diffusion-based conditional generative model and a pre-trained conditional mean estimator, to accurately predict the distribution of $\boldsymbol y$ given $\boldsymbol x$.
1 code implementation • 14 Jun 2022 • Zhendong Wang, Ruijiang Gao, Mingzhang Yin, Mingyuan Zhou, David M. Blei
This paper proposes probabilistic conformal prediction (PCP), a predictive inference algorithm that estimates a target variable by a discontinuous predictive set.
1 code implementation • 14 Jun 2022 • Shentao Yang, Yihao Feng, Shujian Zhang, Mingyuan Zhou
Offline reinforcement learning (RL) extends the paradigm of classical RL algorithms to purely learning from static datasets, without interacting with the underlying environment during the learning process.
3 code implementations • 5 Jun 2022 • Zhendong Wang, Huangjie Zheng, Pengcheng He, Weizhu Chen, Mingyuan Zhou
Both the observed and generated data are diffused by the same adaptive diffusion process.
Ranked #1 on Image Generation on LSUN Bedroom 256 x 256
no code implementations • Findings (NAACL) 2022 • Shujian Zhang, Chengyue Gong, Xingchao Liu, Pengcheng He, Weizhu Chen, Mingyuan Zhou
Active learning, which effectively collects informative unlabeled data for annotation, reduces the demand for labeled data.
2 code implementations • ICLR 2022 • Dongsheng Wang, Dandan Guo, He Zhao, Huangjie Zheng, Korawat Tanwisuth, Bo Chen, Mingyuan Zhou
This paper introduces a new topic-modeling framework where each document is viewed as a set of word embedding vectors and each topic is modeled as an embedding vector in the same embedding space.
1 code implementation • 19 Feb 2022 • Huangjie Zheng, Pengcheng He, Weizhu Chen, Mingyuan Zhou
Employing a forward diffusion chain to gradually map the data to a noise distribution, diffusion-based generative models learn how to generate the data by inferring a reverse diffusion chain.
Ranked #1 on Text-to-Image Generation on CUB
no code implementations • 19 Feb 2022 • Shentao Yang, Zhendong Wang, Huangjie Zheng, Yihao Feng, Mingyuan Zhou
For training more effective agents, we propose a framework that supports learning a flexible yet well-regularized fully-implicit policy.
2 code implementations • 14 Feb 2022 • Huangjie Zheng, Pengcheng He, Weizhu Chen, Mingyuan Zhou
In this paper, to exploit both global and local dependencies without self-attention, we present Mix-Shift-MLP (MS-MLP) which makes the size of the local receptive field used for mixing increase with respect to the amount of spatial shifting.
1 code implementation • 7 Feb 2022 • Yilin He, Chaojie Wang, Hao Zhang, Bo Chen, Mingyuan Zhou
This paper introduces a graph generative process to model how the observed edges are generated by aggregating the node interactions over a set of overlapping node communities, each of which contributes to the edges via a logical OR mechanism.
no code implementations • 15 Dec 2021 • Arman Hasanzadeh, Mohammadreza Armandpour, Ehsan Hajiramezanali, Mingyuan Zhou, Nick Duffield, Krishna Narayanan
By learning distributional representations, we provide uncertainty estimates in downstream graph analytics tasks and increase the expressive power of the predictive model.
no code implementations • NeurIPS 2021 • Mohammadreza Armandpour, Ali Sadeghian, Mingyuan Zhou
The splitting function at each node of CPT is based on the logical disjunction of a community of differently weighted probabilistic linear decision-makers, which also geometrically corresponds to a convex polytope in the covariate space.
1 code implementation • NeurIPS 2021 • Zhibin Duan, Yishi Xu, Bo Chen, Dongsheng Wang, Chaojie Wang, Mingyuan Zhou
Existing deep hierarchical topic models are able to extract semantically meaningful topics from a text corpus in an unsupervised manner and automatically organize them into a topic hierarchy.
1 code implementation • NeurIPS 2021 • Alek Dimitriev, Mingyuan Zhou
Accurately backpropagating the gradient through categorical variables is a challenging task that arises in various domains, such as training discrete latent variable models.
1 code implementation • NeurIPS 2021 • Shujian Zhang, Xinjie Fan, Huangjie Zheng, Korawat Tanwisuth, Mingyuan Zhou
The neural attention mechanism has been incorporated into deep neural networks to achieve state-of-the-art performance in various domains.
1 code implementation • NeurIPS 2021 • Korawat Tanwisuth, Xinjie Fan, Huangjie Zheng, Shujian Zhang, Hao Zhang, Bo Chen, Mingyuan Zhou
Existing methods for unsupervised domain adaptation often rely on minimizing some statistical distance between the source and target samples in the latent space.
no code implementations • ICLR 2022 • Dandan Guo, Long Tian, Minghe Zhang, Mingyuan Zhou, Hongyuan Zha
Since our plug-and-play framework can be applied to many meta-learning problems, we further instantiate it to the cases of few-shot classification and implicit meta generative modeling.
no code implementations • 29 Sep 2021 • Shujian Zhang, Zhibin Duan, Huangjie Zheng, Pengcheng He, Bo Chen, Weizhu Chen, Mingyuan Zhou
Crossformer with states sharing not only provides the desired cross-layer guidance and regularization but also reduces the memory requirement.
no code implementations • 29 Sep 2021 • Shentao Yang, Zhendong Wang, Huangjie Zheng, Mingyuan Zhou
For training more effective agents, we propose a framework that supports learning a flexible and well-regularized policy, which consists of a fully implicit policy and a regularization through the state-action visitation frequency induced by the current policy and that induced by the data-collecting behavior policy.
no code implementations • 29 Sep 2021 • Yilin He, Chaojie Wang, Hao Zhang, Bo Chen, Mingyuan Zhou
In this paper, we introduce a relational graph generative process to model how the observed edges are generated by aggregating the node interactions over multiple overlapping node communities, each of which represents a particular type of relation that contributes to the edges via a logical OR mechanism.
1 code implementation • ACL 2021 • Zhibin Duan, Hao Zhang, Chaojie Wang, Zhengjue Wang, Bo Chen, Mingyuan Zhou
As a result, the backbone learns the shared knowledge among all clusters while modulated weights extract the cluster-specific features.
1 code implementation • 30 Jun 2021 • Zhibin Duan, Dongsheng Wang, Bo Chen, Chaojie Wang, Wenchao Chen, Yewen Li, Jie Ren, Mingyuan Zhou
However, they often assume in the prior that the topics at each layer are independently drawn from the Dirichlet distribution, ignoring the dependencies between the topics both at the same layer and across different layers.
1 code implementation • NeurIPS 2021 • Qizhou Wang, Feng Liu, Bo Han, Tongliang Liu, Chen Gong, Gang Niu, Mingyuan Zhou, Masashi Sugiyama
Reweighting adversarial data during training has been recently shown to improve adversarial robustness, where data closer to the current decision boundaries are regarded as more critical and given larger weights.
no code implementations • 9 Jun 2021 • Shujian Zhang, Xinjie Fan, Bo Chen, Mingyuan Zhou
Attention-based neural networks have achieved state-of-the-art results on a wide range of tasks.
no code implementations • CVPR 2021 • Xinjie Fan, Qifei Wang, Junjie Ke, Feng Yang, Boqing Gong, Mingyuan Zhou
As a generic tool, the improvement introduced by ASR-Norm is agnostic to the choice of ADA methods.
1 code implementation • 28 May 2021 • Alek Dimitriev, Mingyuan Zhou
ARMS uses a copula to generate any number of mutually antithetic samples.
1 code implementation • 10 May 2021 • Dandan Guo, Ruiying Lu, Bo Chen, Zequn Zeng, Mingyuan Zhou
Inspired by recent successes in integrating semantic topics into this task, this paper develops a plug-and-play hierarchical-topic-guided image paragraph generation framework, which couples a visual extractor with a deep topic model to guide the learning of a language model.
1 code implementation • 8 May 2021 • Huangjie Zheng, Xu Chen, Jiangchao Yao, Hongxia Yang, Chunyuan Li, Ya zhang, Hao Zhang, Ivor Tsang, Jingren Zhou, Mingyuan Zhou
We realize this strategy with contrastive attraction and contrastive repulsion (CACR), which makes the query not only exert a greater force to attract more distant positive samples but also do so to repel closer negative samples.
1 code implementation • CVPR 2021 • Mohammadreza Armandpour, Ali Sadeghian, Chunyuan Li, Mingyuan Zhou
We formulate two desired criteria for the space partitioner that aid the training of our mixture of generators: 1) to produce connected partitions and 2) provide a proxy of distance between partitions and data samples, along with a direction for reducing that distance.
Ranked #9 on Image Generation on STL-10
1 code implementation • ICLR 2021 • Xinjie Fan, Shujian Zhang, Korawat Tanwisuth, Xiaoning Qian, Mingyuan Zhou
However, the quality of uncertainty estimation is highly dependent on the dropout probabilities.
1 code implementation • ICLR 2022 • Haoang Chi, Feng Liu, Bo Han, Wenjing Yang, Long Lan, Tongliang Liu, Gang Niu, Mingyuan Zhou, Masashi Sugiyama
In this paper, we demystify assumptions behind NCD and find that high-level semantic features should be shared among the seen and unseen classes.
no code implementations • ICCV 2021 • Yuqi Ding, Yu Ji, Mingyuan Zhou, Sing Bing Kang, Jinwei Ye
Helmholtz stereopsis (HS) exploits the reciprocity principle of light propagation (i. e., the Helmholtz reciprocity) for 3D reconstruction of surfaces with arbitrary reflectance.
no code implementations • 1 Jan 2021 • Ruiying Lu, Bo Chen, Dan dan Guo, Dongsheng Wang, Mingyuan Zhou
Moving beyond conventional Transformers that ignore longer-range word dependencies and contextualize their word representations at the segment level, the proposed method not only captures global semantic coherence of all segments and global word concurrence patterns, but also enriches the representation of each token by adapting it to its local context, which is not limited to the segment it resides in and can be flexibly defined according to the task.
1 code implementation • NeurIPS 2021 • Huangjie Zheng, Mingyuan Zhou
The forward CT is the expected cost of moving a source data point to a target one, with their joint distribution defined by the product of the source probability density function (PDF) and a source-dependent conditional distribution, which is related to the target PDF via Bayes' theorem.
no code implementations • 25 Dec 2020 • Chunyuan Li, Xiujun Li, Lei Zhang, Baolin Peng, Mingyuan Zhou, Jianfeng Gao
Self-supervised pre-training (SSP) employs random image transformations to generate training data for visual representation learning.
Ranked #69 on Self-Supervised Image Classification on ImageNet
no code implementations • NeurIPS 2020 • Chaojie Wang, Hao Zhang, Bo Chen, Dongsheng Wang, Zhengjue Wang, Mingyuan Zhou
To analyze a collection of interconnected documents, relational topic models (RTMs) have been developed to describe both the link structure and document content, exploring their underlying relationships via a single-layer latent representation with limited expressive capability.
1 code implementation • NeurIPS 2020 • Wenchao Chen, Chaojie Wang, Bo Chen, Yicheng Liu, Hao Zhang, Mingyuan Zhou
Incorporating the natural document-sentence-word structure into hierarchical Bayesian modeling, we propose convolutional Poisson gamma dynamical systems (PGDS) that introduce not only word-level probabilistic convolutions, but also sentence-level stochastic temporal transitions.
1 code implementation • 31 Oct 2020 • Ali Lotfi Rezaabad, Rahi Kalantari, Sriram Vishwanath, Mingyuan Zhou, Jonathan Tamir
We show that the existing semi-implicit variational inference objective provably reduces information in the observed graph.
1 code implementation • 21 Oct 2020 • Mohammadreza Armandpour, Mingyuan Zhou
The splitting function at each node of CPT is based on the logical disjunction of a community of differently weighted probabilistic linear decision-makers, which also geometrically corresponds to a convex polytope in the covariate space.
1 code implementation • NeurIPS 2020 • Xinjie Fan, Shujian Zhang, Bo Chen, Mingyuan Zhou
Attention modules, as simple and effective tools, have not only enabled deep neural networks to achieve state-of-the-art results in many domains, but also enhanced their interpretability.
no code implementations • 2 Oct 2020 • Quan Zhang, Huangjie Zheng, Mingyuan Zhou
Leveraging well-established MCMC strategies, we propose MCMC-interactive variational inference (MIVI) to not only estimate the posterior in a time constrained manner, but also facilitate the design of MCMC transitions.
no code implementations • 28 Sep 2020 • Dandan Guo, Bo Chen, Wenchao Chen, Chaojie Wang, Hongwei Liu, Mingyuan Zhou
We develop a recurrent gamma belief network (rGBN) for radar automatic target recognition (RATR) based on high-resolution range profile (HRRP), which characterizes the temporal dependence across the range cells of HRRP.
no code implementations • 28 Sep 2020 • Huangjie Zheng, Mingyuan Zhou
We propose conditional transport (CT) as a new divergence to measure the difference between two probability distributions.
1 code implementation • 25 Jul 2020 • Rahi Kalantari, Mingyuan Zhou
We use the generated random graph, whose number of nonzero-degree nodes is finite, to define both the sparsity pattern and dimension of the latent state transition matrix of a (generalized) linear dynamical system.
3 code implementations • NeurIPS 2020 • Yuguang Yue, Zhendong Wang, Mingyuan Zhou
To improve the sample efficiency of policy-gradient based reinforcement learning algorithms, we propose implicit distributional actor-critic (IDAC) that consists of a distributional critic, built on two deep generator networks (DGNs), and a semi-implicit actor (SIA), powered by a flexible policy distribution.
no code implementations • 15 Jun 2020 • Hao Zhang, Bo Chen, Yulai Cong, Dandan Guo, Hongwei Liu, Mingyuan Zhou
Given a posterior sample of the global parameters, in order to efficiently infer the local latent representations of a document under DATM across all stochastic layers, we propose a Weibull upward-downward variational encoder that deterministically propagates information upward via a deep neural network, followed by a Weibull distribution based stochastic downward generative model.
1 code implementation • 11 Jun 2020 • Mingzhang Yin, Nhat Ho, Bowei Yan, Xiaoning Qian, Mingyuan Zhou
This paper proposes a novel optimization method to solve the exact L0-regularized regression problem, which is also known as the best subset selection.
Methodology
1 code implementation • ICML 2020 • Arman Hasanzadeh, Ehsan Hajiramezanali, Shahin Boluki, Mingyuan Zhou, Nick Duffield, Krishna Narayanan, Xiaoning Qian
We propose a unified framework for adaptive connection sampling in graph neural networks (GNNs) that generalizes existing stochastic regularization methods for training GNNs.
no code implementations • 21 May 2020 • Siamak Zamani Dadaneh, Shahin Boluki, Mingzhang Yin, Mingyuan Zhou, Xiaoning Qian
Semantic hashing has become a crucial component of fast similarity search in many large-scale information retrieval systems, in particular, for text data.
1 code implementation • ICLR 2020 • Liangjian Wen, Yiji Zhou, Lirong He, Mingyuan Zhou, Zenglin Xu
To this end, we propose the Mutual Information Gradient Estimator (MIGE) for representation learning based on the score estimation of implicit distributions.
no code implementations • 12 Feb 2020 • Shahin Boluki, Randy Ardywibowo, Siamak Zamani Dadaneh, Mingyuan Zhou, Xiaoning Qian
In this work, we propose learnable Bernoulli dropout (LBD), a new model-agnostic dropout scheme that considers the dropout rates as parameters jointly optimized with other model parameters.
1 code implementation • 10 Feb 2020 • Yuguang Yue, Yunhao Tang, Mingzhang Yin, Mingyuan Zhou
Reinforcement learning (RL) in discrete action space is ubiquitous in real-world applications, but its complexity grows exponentially with the action-space dimension, making it challenging to apply existing on-policy gradient based deep RL algorithms efficiently.
1 code implementation • ICLR 2020 • Xinjie Fan, Yizhe Zhang, Zhendong Wang, Mingyuan Zhou
To stabilize this method, we adapt to contextual generation of categorical sequences a policy gradient estimator, which evaluates a set of correlated Monte Carlo (MC) rollouts for variance control.
1 code implementation • ICML 2020 • Dandan Guo, Bo Chen, Ruiying Lu, Mingyuan Zhou
To simultaneously capture syntax and global semantics from a text corpus, we propose a new larger-context recurrent neural network (RNN) based language model, which extracts recurrent hierarchical semantic structure via a dynamic deep topic model to guide natural language generation.
1 code implementation • ICLR 2020 • Mingzhang Yin, George Tucker, Mingyuan Zhou, Sergey Levine, Chelsea Finn
If this is not done, the meta-learner can ignore the task training data and learn a single model that performs all of the meta-training tasks zero-shot, but does not adapt effectively to new image classes.
no code implementations • 2 Nov 2019 • Quan Zhang, Qiang Gao, Mingfeng Lin, Mingyuan Zhou
Specifically, we study time to death of three types of lymphoma and show the potential of WDR in modeling nonlinear covariate effects and discovering new diseases.
Survival Analysis Methodology
no code implementations • 1 Nov 2019 • Siamak Zamani Dadaneh, Shahin Boluki, Mingyuan Zhou, Xiaoning Qian
Learning-to-rank methods can generally be categorized into pointwise, pairwise, and listwise approaches.
1 code implementation • ICML 2020 • Zhendong Wang, Mingyuan Zhou
Variational inference is used to approximate the posterior of the local variable, and semi-implicit structure is further introduced to enhance its expressiveness.
1 code implementation • NeurIPS 2019 • Aaron Schein, Scott W. Linderman, Mingyuan Zhou, David M. Blei, Hanna Wallach
This paper presents the Poisson-randomized gamma dynamical system (PRGDS), a model for sequentially observed count tensors that encodes a strong inductive bias toward sparsity and burstiness.
no code implementations • 28 Oct 2019 • Ehsan Hajiramezanali, Arman Hasanzadeh, Nick Duffield, Krishna Narayanan, Mingyuan Zhou, Xiaoning Qian
Stochastic recurrent neural networks with latent random variables of complex dependency structures have shown to be more successful in modeling sequential data than deterministic deep models.
no code implementations • 18 Oct 2019 • Wenyuan Li, Zichen Wang, Yuguang Yue, Jiayun Li, William Speier, Mingyuan Zhou, Corey W. Arnold
In this work, we investigate semi-supervised learning (SSL) for image classification using adversarial training.
no code implementations • 25 Sep 2019 • Dandan Guo, Bo Chen, Ruiying Lu, Mingyuan Zhou
To simultaneously capture syntax and semantics from a text corpus, we propose a new larger-context language model that extracts recurrent hierarchical semantic structure via a dynamic deep topic model to guide natural language generation.
2 code implementations • NeurIPS 2019 • Ehsan Hajiramezanali, Arman Hasanzadeh, Nick Duffield, Krishna R. Narayanan, Mingyuan Zhou, Xiaoning Qian
Representation learning over graph structured data has been mostly studied in static graph settings while efforts for modeling dynamic graphs are still scant.
Ranked #2 on Dynamic Link Prediction on DBLP Temporal
1 code implementation • NeurIPS 2019 • Arman Hasanzadeh, Ehsan Hajiramezanali, Nick Duffield, Krishna R. Narayanan, Mingyuan Zhou, Xiaoning Qian
Compared to VGAE, the derived graph latent representations by SIG-VAE are more interpretable, due to more expressive generative model and more faithful inference enabled by the flexible semi-implicit construction.
no code implementations • 29 May 2019 • Mingzhang Yin, Mingyuan Zhou
To combine explicit and implicit generative models, we introduce semi-implicit generator (SIG) as a flexible hierarchical model that can be trained in the maximum likelihood framework.
1 code implementation • ICLR 2020 • Hao Zhang, Bo Chen, Long Tian, Zhengjue Wang, Mingyuan Zhou
For bidirectional joint image-text modeling, we develop variational hetero-encoder (VHE) randomized generative adversarial network (GAN), a versatile deep generative model that integrates a probabilistic text decoder, probabilistic image encoder, and GAN into a coherent end-to-end multi-modality learning framework.
1 code implementation • 14 May 2019 • Chaojie Wang, Bo Chen, Sucheng Xiao, Mingyuan Zhou
For text analysis, one often resorts to a lossy representation that either completely ignores word order or embeds each word as a low-dimensional dense feature vector.
1 code implementation • 4 May 2019 • Mingzhang Yin, Yuguang Yue, Mingyuan Zhou
To address the challenge of backpropagating the gradient through categorical variables, we propose the augment-REINFORCE-swap-merge (ARSM) gradient estimator that is unbiased and has low variance.
1 code implementation • 2 May 2019 • He Zhao, Piyush Rai, Lan Du, Wray Buntine, Mingyuan Zhou
Many applications, such as text modelling, high-throughput sequencing, and recommender systems, require analysing sparse, high-dimensional, and overdispersed discrete (count-valued or binary) data.
no code implementations • ICLR 2019 • Hao Zhang, Bo Chen, Long Tian, Zhengjue Wang, Mingyuan Zhou
To extract and relate visual and linguistic concepts from images and textual descriptions for text-based zero-shot learning (ZSL), we develop variational hetero-encoder (VHE) that decodes text via a deep probabilisitic topic model, the variational posterior of whose local latent variables is encoded from an image via a Weibull distribution based inference network.
no code implementations • 13 Apr 2019 • Rajat Panda, Ankit Pensia, Nikhil Mehta, Mingyuan Zhou, Piyush Rai
We present a probabilistic framework for multi-label learning based on a deep generative model for the binary label vector associated with each observation.
no code implementations • 9 Apr 2019 • Mingyuan Zhou, Yu Ji, Yuqi Ding, Jinwei Ye, S. Susan Young, Jingyi Yu
In this paper, we introduce a novel concentric multi-spectral light field (CMSLF) design that is able to recover the shape and reflectance of surfaces with arbitrary material in one shot.
no code implementations • 4 Apr 2019 • Zhang Chen, Yu Ji, Mingyuan Zhou, Sing Bing Kang, Jingyi Yu
We avoid the need for spatial constancy of albedo; instead, we use a new measure for albedo similarity that is based on the albedo norm profile.
no code implementations • 13 Mar 2019 • Yunhao Tang, Mingzhang Yin, Mingyuan Zhou
Due to the high variance of policy gradients, on-policy optimization algorithms are plagued with low sample efficiency.
2 code implementations • NeurIPS 2018 • He Zhao, Lan Du, Wray Buntine, Mingyuan Zhou
Recently, considerable research effort has been devoted to developing deep architectures for topic models to learn topic structures.
no code implementations • NeurIPS 2018 • Dandan Guo, Bo Chen, Hao Zhang, Mingyuan Zhou
We develop deep Poisson-gamma dynamical systems (DPGDS) to model sequentially observed multivariate count data, improving previously proposed models by not only mining deep hierarchical latent structure from the data, but also capturing both first-order and long-range temporal dependencies.
no code implementations • NeurIPS 2018 • Ehsan Hajiramezanali, Siamak Zamani Dadaneh, Alireza Karbalayghareh, Mingyuan Zhou, Xiaoning Qian
Second, compared to the number of involved molecules and system complexity, the number of available samples for studying complex disease, such as cancer, is often limited, especially considering disease heterogeneity.
1 code implementation • NeurIPS 2018 • Quan Zhang, Mingyuan Zhou
We propose Lomax delegate racing (LDR) to explicitly model the mechanism of survival under competing risks and to interpret how the covariates accelerate or decelerate the time to event.
1 code implementation • ICLR 2019 • Mingzhang Yin, Mingyuan Zhou
To backpropagate the gradients through stochastic binary layers, we propose the augment-REINFORCE-merge (ARM) estimator that is unbiased, exhibits low variance, and has low computational complexity.
1 code implementation • ICML 2018 • He Zhao, Lan Du, Wray Buntine, Mingyuan Zhou
One important task of topic modeling for text analysis is interpretability.
1 code implementation • ICML 2018 • Mingzhang Yin, Mingyuan Zhou
Semi-implicit variational inference (SIVI) is introduced to expand the commonly used analytic variational distribution family, by mixing the variational parameter with a flexible distribution.
2 code implementations • NeurIPS 2018 • Mingyuan Zhou
Combining Bayesian nonparametrics and a forward model selection strategy, we construct parsimonious Bayesian deep networks (PBDNs) that infer capacity-regularized network architectures from the data and require neither cross-validation nor fine-tuning when training the model.
2 code implementations • NeurIPS 2018 • Bo Han, Jiangchao Yao, Gang Niu, Mingyuan Zhou, Ivor Tsang, Ya zhang, Masashi Sugiyama
It is important to learn various types of classifiers given training data with noisy labels.
Ranked #42 on Image Classification on Clothing1M (using extra training data)
1 code implementation • 22 Mar 2018 • Aaron Schein, Zhiwei Steven Wu, Alexandra Schofield, Mingyuan Zhou, Hanna Wallach
We present a general method for privacy-preserving Bayesian inference in Poisson factorization, a broad class of models that includes some of the most widely used models in the social sciences.
no code implementations • 7 Mar 2018 • Ehsan Hajiramezanali, Siamak Zamani Dadaneh, Paul de Figueiredo, Sing-Hoi Sze, Mingyuan Zhou, Xiaoning Qian
Next-generation sequencing (NGS) to profile temporal changes in living systems is gaining more attention for deriving better insights into the underlying biological mechanisms compared to traditional static sequencing experiments.
1 code implementation • ICLR 2018 • Hao Zhang, Bo Chen, Dandan Guo, Mingyuan Zhou
To train an inference network jointly with a deep generative topic model, making it both scalable to big corpora and fast in out-of-sample prediction, we develop Weibull hybrid autoencoding inference (WHAI) for deep latent Dirichlet allocation, which infers posterior samples via a hybrid of stochastic-gradient MCMC and autoencoding variational Bayes.
no code implementations • 21 Feb 2018 • Rahi Kalantari, Joydeep Ghosh, Mingyuan Zhou
A nonparametric Bayesian sparse graph linear dynamical system (SGLDS) is proposed to model sequentially observed multivariate data.
no code implementations • ICML 2017 • Yulai Cong, Bo Chen, Hongwei Liu, Mingyuan Zhou
It is challenging to develop stochastic gradient based scalable inference for deep discrete latent variable models (LVMs), due to the difficulties in not only computing the gradients, but also adapting the step sizes to different latent factors and hidden layers.
1 code implementation • 19 Jan 2017 • Aaron Schein, Mingyuan Zhou, Hanna Wallach
We introduce a new dynamical system for sequentially observed multivariate count data.
no code implementations • 30 Dec 2016 • Quan Zhang, Mingyuan Zhou
To model categorical response variables given their covariates, we propose a permuted and augmented stick-breaking (paSB) construction that one-to-one maps the observed categories to randomly permuted latent sticks.
1 code implementation • NeurIPS 2016 • Aaron Schein, Hanna Wallach, Mingyuan Zhou
This paper presents a dynamical system based on the Poisson-Gamma construction for sequentially observed multivariate count data.
no code implementations • 23 Aug 2016 • Mingyuan Zhou
To construct flexible nonlinear predictive distributions, the paper introduces a family of softplus function based regression models that convolve, stack, or combine both operations by convolving countably infinite stacked gamma distributions, whose scales depend on the covariates.
1 code implementation • 6 Jun 2016 • Aaron Schein, Mingyuan Zhou, David M. Blei, Hanna Wallach
We introduce Bayesian Poisson Tucker decomposition (BPTD) for modeling country--country interaction event data.
no code implementations • CVPR 2016 • Nianyi Li, Haiting Lin, Bilin Sun, Mingyuan Zhou, Jingyi Yu
In this paper, we present a novel LF sampling scheme by exploiting a special non-centric camera called the crossed-slit or XSlit camera.
no code implementations • 25 Apr 2016 • Mingyuan Zhou
A common approach to analyze a covariate-sample count matrix, an element of which represents how many times a covariate appears in a sample, is to factorize it under the Poisson likelihood.
no code implementations • 30 Dec 2015 • Ayan Acharya, Joydeep Ghosh, Mingyuan Zhou
A gamma process dynamic Poisson factor analysis model is proposed to factorize a dynamic count matrix, whose columns are sequentially observed count vectors.
no code implementations • 9 Dec 2015 • Mingyuan Zhou, Yulai Cong, Bo Chen
To infer multilayer deep representations of high-dimensional discrete and nonnegative real vectors, we propose an augmentable gamma belief network (GBN) that factorizes each of its hidden layers into the product of a sparse connection weight matrix and the nonnegative real hidden units of the next layer.
no code implementations • NeurIPS 2015 • Mingyuan Zhou, Yulai Cong, Bo Chen
Example results on text analysis illustrate interesting relationships between the width of the first layer and the inferred network structure, and demonstrate that the PGBN, whose hidden units are imposed with correlated gamma priors, can add more layers to increase its performance gains over Poisson factor analysis, given the same limit on the width of the first layer.
no code implementations • 25 Jan 2015 • Mingyuan Zhou
A hierarchical gamma process infinite edge partition model is proposed to factorize the binary adjacency matrix of an unweighted undirected relational network under a Bernoulli-Poisson link.
no code implementations • NeurIPS 2014 • Mingyuan Zhou
The beta-negative binomial process (BNBP), an integer-valued stochastic process, is employed to partition a count vector into a latent random count matrix.
no code implementations • 28 Oct 2014 • Mingyuan Zhou
The beta-negative binomial process (BNBP), an integer-valued stochastic process, is employed to partition a count vector into a latent random count matrix.
no code implementations • 12 Apr 2014 • Mingyuan Zhou, Oscar Hernan Madrid Padilla, James G. Scott
We define a family of probability distributions for random count matrices with a potentially unbounded number of rows and columns.
no code implementations • 7 Oct 2013 • Mingyuan Zhou
The paper introduces the concept of a cluster structure to define a joint distribution of the sample size and its exchangeable random partitions.
no code implementations • NeurIPS 2012 • Mingyuan Zhou, Lawrence Carin
By developing data augmentation methods unique to the negative binomial (NB) distribution, we unite seemingly disjoint count and mixture models under the NB process framework.
1 code implementation • 15 Sep 2012 • Mingyuan Zhou, Lawrence Carin
A gamma process is employed to model the rate measure of a Poisson process, whose normalization provides a random probability measure for mixture modeling and whose marginalization leads to an NB process for count modeling.
no code implementations • NeurIPS 2009 • Mingyuan Zhou, Haojun Chen, Lu Ren, Guillermo Sapiro, Lawrence Carin, John W. Paisley
The beta process is employed as a prior for learning the dictionary, and this non-parametric method naturally infers an appropriate dictionary size.