About Me

 

I am an assistant professor in the Department of Computer Science at Rutgers University, where I direct a Machine Learning Lab. Previously I was a Postdoctoral Research Associate at Computer Science & Artificial Intelligence Lab of Massachusetts Institute of Technology, working with Prof. Dina Katabi and Prof. Tommi Jaakkola. I obtained Ph.D degree in CSE department, Hong Kong University of Science and Technology. My supervisor was Prof. Dit-Yan Yeung. I was a visiting scholar working with Prof. Eric Xing's group in Machine Learning department of Carnegie Mellon University. I am also a Microsoft Fellow and received the Baidu Research Fellowship. Before my Ph.D, I got my BS degree from Shanghai Jiao Tong University, 2013 under the supervision of Prof. Wu-Jun Li.

Email: hoguewang AT gmail.com / hw488 AT cs.rutgers.edu / hogue.wang AT rutgers.edu / hwang87 AT mit.edu

Recruiting: I am recruiting PhD students starting from Fall 2025 as well as interns. Send me an email if you are interested in working with me at Rutgers.

[Twitter] [Facebook] [Github] [LinkedIn] [Medium] [CV]

Research Interest

 

My research interest focuses on statistical machine learning and deep learning. Currently, I mainly work on Bayesian deep learning, probabilistic methods, game-theoretic approaches, and their applications in trustworthy & safe AI (interpretability, robustness, alignment, etc.), healthcare, recommender systems, computer vision (including LMM), natural language processing (including LLM), network analysis, and data mining.

Updates

 

  • Our papers on Bayesian large language models and natural counterfactual inference, "BLoB: Bayesian Low-Rank Adaptation by Backpropagation for Large Language Models" and "Natural Counterfactuals With Necessary Backtracking" are accepted at NeurIPS (9/25/24).

  • Our paper on Bayesian deep learning for interpretable large language models, "Variational Language Concepts for Interpreting Foundation Language Models", is accepted at Findings of EMNLP (9/21/24).

  • Grateful to receive the NSF CAREER Award on Robustifying AI with Bayesian Deep Learning (07/10/24).

  • Grateful to receive an NIH R01 Award as PI, "Counterfactual Explanations for AI-Assisted Cancer Diagnosis and Subtyping" (07/01/24).

  • Our paper on large language models for medical education, "Benchmarking Large Language Models on Communicative Medical Coaching: A Dataset and a Novel System", is accepted at Findings of ACL (5/16/24).

  • Our papers on safe & trustworthy large language models and Bayesian deep learning, "Probabilistic Conceptual Explainers: Towards Trustworthy Conceptual Explanations for Vision Foundation Models" and "Delving into Differentially Private Transformer" are accepted at ICML (5/1/24).

  • We are organizing the ICML 2024 Workshop on "Foundation Models in the Wild" (3/27/24).

  • Our paper on safe & trustworthy large language models, "LatticeGen: Hiding Generated Text in a Lattice for Privacy-Aware Large Language Model Generation on Cloud", is accepted at Findings of NAACL (3/13/24).

  • Grateful to receive the Microsoft Research AI & Society Fellowship (03/01/24).

  • Our papers on safe & trustworthy large language models, Bayesian deep learning, domain adaptation, and interpretability, "Detecting Text from Large Language Models via Rewriting", "Energy-Based Concept Bottleneck Models: Unifying Prediction, Concept Intervention, and Probabilistic Interpretations", "Continuous Invariance Learning" are accepted at ICLR (1/16/24).

  • Our paper on multi-domain active learning and domain adaptation, "Composite Active Learning: Towards Multi-Domain Active Learning with Theoretical Guarantees", is accepted at AAAI (11/9/23).

  • Our papers on Bayesian deep learning, domain adaptation, and continual learning, "A Unified Approach to Domain Incremental Learning with Memory: Theory and Algorithm", and "Variational Imbalanced Regression: Fair Uncertainty Quantification via Probabilistic Smoothing" are accepted at NeurIPS (9/21/23).

  • Our paper on learning optimization landscape, "Landscape Learning for Neural Network Inversion", is accepted at ICCV (7/14/23).

  • Our papers on Bayesian deep learning, causality, interpretable ML, robustness, and domain adaptation, "Self-Interpretable Time Series Prediction with Counterfactual Explanations", "Taxonomy-Structured Domain Adaptation", and "Robust Perception through Equivariance" are accepted at ICML (4/24/23).

  • Our paper on Bayesian deep learning for domain adaptation, "Domain-Indexing Variational Bayes for Domain Adaptation", is accepted at ICLR (1/20/23).

  • Our Nature Medicine paper on machine learning for health, "Artificial Intelligence-Enabled Detection and Assessment of Parkinson’s Disease Using Nocturnal Breathing Signals", has been selected as one of the Ten Notable Advances in 2022 by Nature Medicine (1/10/23).

  • We are organizing the CVPR 2023 Workshop on "New Frontiers in Visual Language Reasoning: Compositionality, Prompts and Causality" (12/15/22).

  • Our paper on Bayesian deep learning for speech recognition and education, "Unsupervised Mismatch Localization in Cross-Modal Sequential Data with Application to Mispronunciations Localization", is accepted at TMLR (12/8/22).

  • Our paper on Bayesian deep learning for federated learning, "FedNP: Towards Non-IID Federated Learning via Federated Neural Propagation", is accepted at AAAI (11/18/22).

  • Our paper on causal and counterfactual recommender systems, "Collaborative Counterfactual Reasoning", is accepted at WSDM (10/18/22).

  • Our papers on Bayesian deep learning, continuously streaming domain adaptation, and spatio-temporal forecasting, "Extrapolative Continuous-Time Bayesian Neural Network for Fast Training-Free Test-Time Adaptation" and "Earthformer: Exploring Space-Time Transformers for Earth System Forecasting" are accepted at NeurIPS (09/14/22).

  • Our paper on multi-domain imbalanced learning and deep learning for health, "Artificial Intelligence-Enabled Detection and Assessment of Parkinson’s Disease Using Nocturnal Breathing Signals", is accepted at Nature Medicine (8/2/22).

  • Our papers on multi-domain imbalanced learning and relational forecasting, "On Multi-Domain Long-Lailed Recognition, Generalization and Beyond" and "Social ODE: Multi-Agent Trajectory Forecasting with Neural Ordinary Differential Equations" are accepted at ECCV (07/03/22).

  • Our paper, "OrphicX: A Causality-Inspired Latent Variable Model for Interpreting Graph Neural Networks", is a best paper finalist at CVPR 2022 (6/24/22).

  • Our paper on domain adaptation Transformer, "Domain Adaptation for Time Series Forecasting via Attention Sharing", is accepted at ICML (05/13/22).

  • Our paper on Bayesian deep learning and interpretable ML for healthcare, "'My Nose is Running.' 'Are you Also Coughing?': Building a Medical Diagnosis Agent with Interpretable Inquiry Logics", is accepted at IJCAI (04/20/22).

  • Our three papers on causality, interpretable ML and Bayesian deep learning, "Causal Transportability for Visual Recognition", "OrphicX: A Causality-Inspired Latent Variable Model for Interpreting Graph Neural Networks", and "Bayesian Invariant Risk Minimization", are accepted at CVPR (03/03/22).

  • Our paper, "Graph-Relational Domain Adaptation", is accepted at ICLR (1/20/22).

  • We are organizing the ICLR 2022 Workshop on "PAIR^2Struct: Privacy, Accountability, Interpretability, Robustness, Reasoning on Structured Data" (12/06/21).

  • Our two papers on uncertainty estimation, "Context Uncertainty in Contextual Bandits with Applications to Recommender Systems" and "Training-Free Uncertainty Estimation for Dense Regression: Sensitivity as a Surrogate", are accepted at AAAI (12/01/21).

  • Grateful to receive NSF grant IIS-2127918 as PI, "RI: Small: Enabling Interpretable AI via Bayesian Deep Learning" (08/25/21).

  • Our two papers, "Adversarial Attacks Are Reversible with Natural Supervision" and "Paint Transformer: Feed Forward Neural Painting with Stroke Prediction", are accepted at ICCV (07/22/21).

  • Our three papers, "STRODE: Stochastic Boundary Ordinary Differential Equation", "Correcting Exposure Bias for Link Recommendation", and "Delving into Deep Imbalanced Regression", are accepted at ICML (05/08/21).

  • Received Amazon Faculty Research Award to work on Domain Adaptation, Recommender Systems, Forecasting, and Bayesian Deep Learning. (04/28/21).

  • Our paper on causal learning and deep learning, "Generative Interventions for Causal Learning", is accepted at CVPR (02/28/21).

  • Our paper on AI and Bayesian deep learning for health, "Assessment of Medication Self-Administration Using Artificial Intelligence", is accepted at Nature Medicine (10/30/20).

  • Our Bayesian deep learning survey paper, "A Survey on Bayesian Deep Learning", is accepted and published at ACM Computing Surveys (10/01/20).

  • Our work BodyCompass was covered by: MIT News, Engadget, Yahoo, Technology Networks, Sleep Review, TechTimes, and other media outlets (09/30/20).

  • Our work on COVID-19 patient monitoring was covered by: CSAIL news, TechCrunch, Engadget, and other media outlets (09/30/20).

  • Our two papers, "Continuously Indexed Domain Adaptation" and "Deep Graph Random Process for Relational-Thinking-Based Speech Recognition", are accepted at ICML (06/06/20).

  • We released a new TensorFlow implementation for our KDD 2015 paper "Collaborative deep learning for recommender systems" (06/06/20).

  • Our paper, "Learning Guided Electron Microscopy with Active Acquisition", is accepted at MICCAI(06/06/20).

  • A new project page for an ongoing survey on Bayesian Deep Learning is set up (08/30/19).

  • A new project page for our NPN paper is set up with both Matlab and PyTorch code (06/30/19).

  • We are organizaing the ICML 2019 Workshop on "Learning and Reasoning with Graph-Structured Representations" (15/03/19).

  • We are organizaing the CVPR 2019 Workshop on "Towards Causal, Explainable and Universal Medical Visual Diagnosis" (03/11/19).

  • Our paper, "Rethinking Knowledge Graph Propagation for Zero-Shot Learning", is accepted at CVPR (02/24/19).

  • Our work on Deep Bayesian Networks is reported by MIT News (01/25/19).

  • Our paper, "ProbGAN: Towards Probabilistic GAN with Theoretical Guarantees", is accepted at ICLR (12/22/18).

  • Our two papers, "Bidirectional Inference Networks: A Class of Deep Bayesian Networks for Health Profiling" and "Recurrent Poisson process unit for speech recognition", are accepted at AAAI (11/01/18).

  • Our paper, "Deep Learning for Precipitation Nowcasting: A Benchmark and a New Model", is accepted at NIPS (09/05/17).

  • A new project page for our CDL paper is set up with a brief list of CDL variants (06/12/17).

  • Our paper, "Relational Deep Learning: A Deep Latent Variable Model for Link Prediction", is accepted at AAAI (11/11/16).

  • Our survey/review paper on Bayesian deep learning, "Towards Bayesian Deep Learning: A Framework and Some Existing Methods", is accepted in TKDE (08/22/16).

  • Two of our papers, "Natural Parameter Networks: A Class of Probabilistic Neural Networks" and "Collaborative Recurrent Autoencoder: Recommend While Learning to Fill in the Blanks", are accepted at NIPS (08/15/16).

  • Give the talk "Bayesian Deep Learning for Integrated Intelligence: Bridging the Gap between Perception and Inference" at the Chinese University of Hong Kong (06/17/16). [slides]

  • Give the talk "Bayesian Deep Learning for Integrated Intelligence: Bridging the Gap between Perception and Inference" at the Baidu NLP Seminar (06/13/16). [slides]

  • We gave a talk about Bayesian Deep Learning at ACML (11/22/15). [slides]

  • Give a talk about Bayesian Deep Learning at MSRA (09/11/15) and Baidu (11/05/15). [slides]

  • Our paper "Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting" is accepted at NIPS. (09/04/15) [pdf]

  • Our paper "Collaborative Deep Learning for Recommender Systems" is accepted at SIGKDD. (05/13/15) [pdf]

  • Give a talk about " Relational Stacked Denoising Autoencoder for Tag Recommendation" at HKUST-EPFL Workshop on Data Science. (12/02/14) [slides]

  • This homepage is set up. (11/18/14)

 
hit counters free