Zhi Zhang

PhD Candidate@UCLA. Reinforcement Learning, LLM Post-Training, Agentic AI.

personal.jpg

👋 Hi! I am Zhi Zhang (Zach), a PhD candidate in the Department of Statistics and Data Science at UCLA. I’m fortunate to be advised by Prof. Arash Amini.

Previously, I conducted an Applied Scientist internship at AWS AI (Summer 2025), where I developed AERO (Adaptive Efficient Rollout Optimization) for RL-based LLM fine-tuning, and an AI Research internship at eBay (Summer 2025), where I built ReflexAgent for agentic NER. I hold degrees from Northwestern University (Ph.D. in CS, completed with M.S.), UC Davis (M.S. in Statistics), and Georgia Tech (M.S. in CS).

My research focuses on LLM post-training and RL fine-tuning (GRPO, PPO, RLHF), compute efficiency optimization, Agentic AI systems, and multi-agent reinforcement learning. I’m particularly interested in:

  • RL Post-Training for LLMs: Developing efficient algorithms for reinforcement learning-based fine-tuning of large language models
  • LLM Efficiency: Reducing computational costs while maintaining or improving model performance
  • Agentic AI: Building intelligent agents that can reason, plan, and execute complex tasks
  • Multi-Agent RL: Designing algorithms for cooperative and competitive multi-agent systems

I have published at top venues including ICLR (Spotlight), AISTATS, NeurIPS, and AAMAS, and serve as a reviewer for ICML, NeurIPS, AISTATS, and ICLR.

News

Jan 26, 2026 Two papers accepted to ICLR 2026! 🎉
Oct 1, 2025 🔬 Completed Applied Scientist internship at AWS AI! Developed AERO for efficient RL-based LLM fine-tuning.
Jan 15, 2025 📄 Two papers accepted to AISTATS 2025: Quantile Additive Trend Filtering and Lifelong RL with PAC-Bayes!

Selected Publications

  1. Preprint
    Train Less, Learn More: Adaptive and Efficient Rollout Optimization for Group-Based Reinforcement Learning
    Zhi Zhang, Zhen Han, and  others
    2026
  2. arXiv
    Flow Matching Generalizes Through Discretization Bias
    Zhixin Zhou, Zhi Zhang, and Arash Amini
    arXiv preprint 2026
    PDF
  3. ICLR
    EDIVAL-Agent: An Object-Centric Framework for Automated, Fine-Grained Evaluation of Multi-Turn Editing
    Tianyu Chen, Yasi Zhang, Zhi Zhang, and  others
    In International Conference on Learning Representations 2026
    PDF
  4. ICLR
    Single Index Bandits: Generalized Linear Contextual Bandits with Unknown Reward Functions
    Yue Kang, Mingshuo Liu, Bongsoo Yi, Jing Lyu, Zhi Zhang, Doudou Zhou, and Yao Li
    In International Conference on Learning Representations 2026
    PDF
  5. AISTATS
    Quantile Additive Trend Filtering
    Zhi Zhang, Kyle Ritscher, and Oscar Madrid Padilla
    In International Conference on Artificial Intelligence and Statistics 2025
    PDF
  6. arXiv
    Dense ReLU Neural Network for Temporal-Spatial Model
    Carlos Padilla, Zhi Zhang, Oscar Padilla, and Daren Wang
    arXiv preprint 2025
    PDF
  7. AISTATS
    Statistical Guarantees for Lifelong Reinforcement Learning Using PAC-Bayes Theory
    Zhi Zhang, Chris Chow, Yasi Zhang, and  others
    In International Conference on Artificial Intelligence and Statistics 2025
    PDF
  8. ICLR
    Reinforcement Learning under a Multi-agent Predictive State Representation Model: Method and Theory
    Zhi Zhang, Zhuoran Yang, Han Liu, Pratap Tokekar, and Furong Huang
    In International Conference on Learning Representations 2022
    PDF
  9. AAMAS
    Integrating Independent and Centralized Multi-agent Reinforcement Learning for Traffic Signal Network Optimization
    Zhi Zhang, Jiachen Yang, and Hongyuan Zha
    In International Conference on Autonomous Agents and Multi-Agent Systems 2020
    PDF
  10. AISTATS
    Multivariate Time Series Forecasting by Graph Attention Networks with Theoretical Guarantees
    Zhi Zhang, Weijian Li, and Han Liu
    In International Conference on Artificial Intelligence and Statistics 2024
    PDF

Education

  • University of California, Los Angeles, 2022.09 - Present
    PhD in Statistics and Data Science
  • Northwestern University, 2020 - 2022
    PhD in Computer Science (completed with M.S.)

Experience

  • Amazon AWS AI Labs, Applied Scientist Intern, Summer 2025
  • eBay, AI Research Intern, Summer 2024