Dingyang Chen
I am an Applied Scientist at Amazon. I received my Ph.D. in Computer Science at the University of South Carolina, advised by Prof. Qi Zhang. I also hold a Master of Engineering from UC Berkeley and a B.S. from UIUC.
My research focuses on leveraging structural properties, such as Euclidean equivariance, in reinforcement learning to improve efficiency. Recently, I have focused on adapting large language models to sequential decision-making tasks.
Professional Experience
Amazon
Jan 2025 - PresentApplied Scientist, Seattle, WA
- Fine-tuned LLMs to detect and assess defects in Amazon’s customer service chatbot responses.
Amazon
May 2024 - Aug 2024Applied Scientist Intern, Bellevue, WA
- Developed multi-armed bandit models for cost-of-goods prediction.
Siemens
May 2022 - Aug 2022Research Intern, Remote
- Applied reinforcement learning algorithms to optimize power system reconfiguration.
Publications
-
Correlated Policy Optimization in Multi-Agent Subteams
International Conference on Learning Representations (ICLR), 2026
-
Reinforcement Learning with Euclidean Data Augmentation for State-Based Continuous Control
Neural Information Processing Systems (NeurIPS), 2024
-
Convergence Rates of Bayesian Network Policy Gradient for Cooperative Multi-Agent Reinforcement Learning
Workshop on Bayesian Decision-making and Uncertainty at NeurIPS, 2024
-
Efficient Sequential Decision Making with Large Language Models
Empirical Methods in Natural Language Processing (EMNLP), 2024
-
E(3)-Equivariant Actor-Critic Methods for Cooperative Multi-Agent Reinforcement Learning
International Conference on Machine Learning (ICML), 2024
-
Context-Aware Bayesian Network Actor-Critic Methods for Cooperative Multi-Agent Reinforcement Learning
International Conference on Machine Learning (ICML), 2023
-
Communication-Efficient Actor-Critic Methods for Homogeneous Markov Games
International Conference on Learning Representations (ICLR), 2022
-
Convergence and Price of Anarchy Guarantees of the Softmax Policy Gradient in Markov Potential Games
Decision Awareness in Reinforcement Learning Workshop at ICML, 2022
-
A Meta-Gradient Approach to Learning Cooperative Multi-Agent Communication Topology
Workshop on Meta-Learning at NeurIPS, 2021
Professional Service
Conference Reviewer / Program Committee
- ICML 2024, 2025, 2026
- AISTATS 2026
- AAAI 2025
- ICLR 2024
- NeurIPS 2023
- NeurIPS 2022 Deep RL Workshop
Journal Reviewer
- IEEE Transactions on Neural Networks and Learning Systems (TNNLS)
Teaching Experience
Graduate Teaching Assistant
- Visualization Tools, University of South Carolina (Fall 2024)
- Algorithmic Design I, University of South Carolina (Fall 2023)