My picture

He Zhu

Assistant Professor
Computer Science Department
Rutgers University, New Brunswick
Email: hz375@cs.rutgers.edu


[Awards] [Service] [Teaching] [Publications]
Multiple positions regarding Neurosymbolic Programming and Program Synthesis are available in my group for Fall 2023!
Please drop me an email with your CV if you are interested. Application deadline is Jan. 1st. 2023 for PhD students.
I am an assistant professor in the Department of Computer Science at Rutgers University-New Brunswick. My research focuses on neurosymbolic programming, which lies at the intersection of automated programming and deep learning and spans over programming languages, formal methods, and machine learning. I investigate program synthesis and formal program reasoning techniques to make machine learning systems more reliable and trustworthy. The long-term research goal is to build intelligent and interpretable AI systems that allow the tight integration of deep learning and symbolic reasoning and that can be certified robust and reliable.

I obtained my Ph.D. from Purdue CS, advised by Suresh Jagannathan.

Publications

Instructing Goal-Conditioned Reinforcement Learning Agents with Temporal Logic Objectives [To Appear]
Wenjie Qiu, Wensen Mao and He Zhu.
Neural Information Processing Systems (NeurIPS), 2023

Verification-guided Programmatic Controller Synthesis [pdf]
Yuning Wang and He Zhu.
29th International Conference on Tools and Algorithms for the Construction and Analysis of Systems (TACAS), 2023

ReLAX: Reinforcement Learning Agent Explainer for Arbitrary Predictive Models [pdf]
Ziheng Chen, Fabrizio Silvestri, Jia Wang, He Zhu, Hongshik Ahn and Gabriele Tolomei.
Proceedings of the 31st ACM International Conference on Information and Knowledge Management (CIKM), 2022

Learn Basic Skills and Reuse: Modularized Adaptive Neural Architecture Search (MANAS) [pdf]
Hanxiong Chen, Yunqi Li, Jia Wang, He Zhu and Yongfeng Zhang.
Proceedings of the 31st ACM International Conference on Information and Knowledge Management (CIKM), 2022

Defending Observation Attacks in Deep Reinforcement Learning via Detection and Denoising [pdf]
Zikang Xiong, Joe Eappen, He Zhu and Suresh Jagannathan.
European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD), 2022

Programmatic Reinforcement Learning without Oracles [pdf]
Spotlight
Wenjie Qiu and He Zhu.
International Conference on Learning Representations (ICLR), 2022

Graph Collaborative Reasoning [pdf]
Hanxiong Chen, Yunqi Li, Shaoyun Shi, Shuchang Liu, He Zhu and Yongfeng Zhang.
Proceedings of the 15th ACM International Conference on Web Search and Data Mining (WSDM), 2022.

Differentiable Synthesis of Program Architectures [pdf]
Guofeng Cui and He Zhu.
Neural Information Processing Systems (NeurIPS), 2021

ART: Abstraction Refinement-Guided Training for Provably Correct Neural Networks [pdf]
Xuankang Lin, He Zhu, Roopsha Samanta and Suresh Jagannathan.
Formal Methods in Computer-Aided Design (FMCAD), 2020

An Inductive Synthesis Framework for Verifiable Reinforcement Learning [pdf][code]
ACM SIGPLAN distinguished paper award
He Zhu, Zikang Xiong, Stephen Magill and Suresh Jagannathan
Proceedings of the 40th ACM SIGPLAN conference on Programming Language Design and Implementation (PLDI), 2019

A Data-Driven CHC Solver [pdf][code]
ACM SIGPLAN distinguished paper award
He Zhu, Stephen Magill and Suresh Jagannathan
Proceedings of the 39th ACM SIGPLAN conference on Programming Language Design and Implementation (PLDI), 2018

Automatically Learning Shape Specifications [pdf][code]
He Zhu, Gustavo Petri and Suresh Jagannathan
Proceedings of the 37th ACM SIGPLAN conference on Programming Language Design and Implementation (PLDI), 2016

Learning Refinement Types [pdf][code]
He Zhu, Aditya V. Nori and Suresh Jagannathan
Proceedings of the 20th ACM SIGPLAN International Conference on Functional Programming (ICFP), 2015

Poling: SMT Aided Linearizability Proofs [pdf]
He Zhu, Gustavo Petri and Suresh Jagannathan
Proceedings of the 27th International Conference on Computer Aided Verification (CAV), 2015

Dependent Array Type Inference from Tests [pdf]
He Zhu, Aditya V. Nori and Suresh Jagannathan
Proceedings of the 16th International Conference on Verification, Model Checking, and Abstract Interpretation (VMCAI), 2015

Compositional and Lightweight Dependent Type Inference for ML [pdf]
He Zhu and Suresh Jagannathan
Proceedings of the 14th International Conference on Verification, Model Checking, and Abstract Interpretation (VMCAI), 2013

Students

Guofeng Cui, Wenjie Qiu, Yuning Wang, Wensen Mao, Zining Fan, Yuanlin Duan.

Acknowledgement

My research is supported by:
  • NSF Grant (2021): Synthesis and Verification for Programmatic Reinforcement Learning
  • NSF Grant (2020): Formal Symbolic Reasoning of Deep Reinforcement Learning Systems
  • DARPA Grant (2020): Symbiotic Design for Cyber Physical Systems

    • Teaching

  • CS 314. Principles of Programming Languages [F23] [S22] [S21] [S20]
  • CS 515. Programming Languages and Compilers [F22] [F21] [F20] [F19]

    • Service

  • 2024: ICLR Reviewer
  • 2023: PLDI PC, NeurIPS Reviewer, NSF Panelist
  • 2022: OOPSLA PC, ICML Reviewer, FMSD Reviewer, NSF Panelist
  • 2021: PLDI PC, HCVS PC, TSE Reviewer
  • 2020: CAV AEC co-Chair, CAV PC, NSF Panelist, TSE Reviewer
  • 2019: HCVS PC, TSE Reviewer

    • Awards

  • ACM SIGPLAN PLDI 2019 Distinguished Paper Award.
  • ACM SIGPLAN PLDI 2018 Distinguished Paper Award.
  • Maurice H. Halstead Memorial Award (Purdue University).