Publications

Gradient Estimation for Exactly-k Constraints

NeurIPS 2023 Workshop on AI for Scientific Discovery
Ruoyan Li, Dipti Ranjan Sahu, Guy Van den Broeck, Zhe Zeng

Collapsed Inference for Bayesian Deep Learning

ICML 2023 Workshop on Structured Probabilistic Inference & Generative Modeling     Oral Presentation (Top 5%)
Zhe Zeng, Guy Van den Broeck

Collapsed Inference for Bayesian Deep Learning

NeurIPS 2023     Oral Presentation at ICML 2023 Workshop on SPIGM (Top 5%)
Zhe Zeng, Guy Van den Broeck
PDF Code
Collapsed Inference for Bayesian Deep Learning

A Unified Approach to Count-Based Weakly-Supervised Learning

NeurIPS 2023
Vinay Shukla, Zhe Zeng, Kareem Ahmed, Guy Van den Broeck
PDF Code
A Unified Approach to Count-Based Weakly-Supervised Learning

Probabilistic Rewired Message-passing Neural Networks

arXiv
Chendi Qian, Andrei Manolache, Kareem Ahmed, Zhe Zeng, Guy Van den Broeck, Mathias Niepert, Christopher Morris
PDF

SIMPLE: A Gradient Estimator for k-Subset Sampling

ICML 2023 Workshop on Differentiable Almost Everything
Kareem Ahmed, Zhe Zeng, Mathias Niepert, Guy Van den Broeck

Probabilistic Task-Adaptive Graph Rewiring

ICML 2023 Workshop on Differentiable Almost Everything
Chendi Qian, Andrei Manolache, Kareem Ahmed, Zhe Zeng, Guy Van den Broeck, Mathias Niepert, Christopher Morris
PDF
Probabilistic Task-Adaptive Graph Rewiring

A Unified Approach to Count-Based Weakly-Supervised Learning

ICML 2023 Workshop on Differentiable Almost Everything
Vinay Shukla, Zhe Zeng, Kareem Ahmed, Guy Van den Broeck

A Unified Knowledge Distillation Framework for Deep Directed Graphical Models

CVPR 2023
Yizhuo Chen, Kaizhao Liang, Zhe Zeng, Yifei Yang, Shuochao Yao, Huajie Shao
PDF
A Unified Knowledge Distillation Framework for Deep Directed Graphical Models

SIMPLE: A Gradient Estimator for k-Subset Sampling

ICLR 2023
Kareem Ahmed, Zhe Zeng, Mathias Niepert, Guy Van den Broeck
PDF Code
SIMPLE: A Gradient Estimator for k-Subset Sampling

SIMPLE: A Gradient Estimator for k-Subset Sampling

SoCal NLP Symposium 2022
Kareem Ahmed, Zhe Zeng, Mathias Niepert, Guy Van den Broeck

Tractable Computation of Expected Kernels

UAI 2021
Wenzhe Li, Zhe Zeng, Antonio Vergari, Guy Van den Broeck
PDF Code Poster Slides Video
Tractable Computation of Expected Kernels

Is Parameter Learning via Weighted Model Integration Tractable?

TPM Workshop at UAI 2021
Zhe Zeng, Paolo Morettin, Fanqi Yan, Antonio Vergari, Guy Van den Broeck
Slides
Is Parameter Learning via Weighted Model Integration Tractable?

Probabilistic Inference with Algebraic Constraints: Theoretical Limits and Practical Approximations

NeurIPS 2020     Spotlight presentation, acceptance rate 280/9454 = 2.96%
Zhe Zeng, Paolo Morettin, Fanqi Yan, Antonio Vergari, Guy Van den Broeck
PDF Code Poster Slides Video
Probabilistic Inference with Algebraic Constraints: Theoretical Limits and Practical Approximations

Relax, Compensate and then Integrate

DeCoDeML Workshop at ECML-PKDD 2020
Zhe Zeng, Paolo Morettin, Fanqi Yan, Antonio Vergari, Guy Van den Broeck
Code Slides Video
Relax, Compensate and then Integrate

Scaling up Hybrid Probabilistic Inference with Logical and Arithmetic Constraints via Message Passing

ICML 2020
Zhe Zeng, Paolo Morettin, Fanqi Yan, Antonio Vergari, Guy Van den Broeck
PDF Code Slides Video
Scaling up Hybrid Probabilistic Inference with Logical and Arithmetic Constraints via Message Passing

Hybrid Probabilistic Inference with Logical Constraints: Tractability and Message Passing

KR2ML Workshop at NeurIPS 2019
Zhe Zeng, Fanqi Yan, Paolo Morettin, Antonio Vergari, Guy Van den Broeck
Code Poster Slides
Hybrid Probabilistic Inference with Logical Constraints: Tractability and Message Passing

Efficient Search-Based Weighted Model Integration

IJCAI 2019 Workshop on Declarative Learning Based Programming
Zhe Zeng, Guy Van den Broeck

Efficient Search-Based Weighted Model Integration

UAI 2019
Zhe Zeng, Guy Van den Broeck
PDF Code Poster Slides
Efficient Search-Based Weighted Model Integration

Stein Variational Message Passing for Continuous Graphical Models

ICML 2018
Dilin Wang, Zhe Zeng, Qiang Liu
PDF Code Slides Video
Stein Variational Message Passing for Continuous Graphical Models