Paria Rashidinejad

Postdoctoral Scholar

University of California, Berkeley

About

I’m a Postdoctoral Scholar at Berkeley AI Research (BAIR) Lab and the Center for Human-Compatible AI (CHAI). I recently received my Ph.D. in EECS from UC Berkeley advised by Stuart Russell and Jiantao Jiao .

My research is on mathematical foundations of AI and designing capable and general-purpose AI systems for reliable integration into the real world. My recent research has been focusing on reinforcement learning and machine learning algorithms for autonomous inference, prediction, and decision-making. I also work on machine learning applications in areas such as healthcare, robotics, autonomous driving, and systems.

Interests

Artificial Intelligence
Reinforcement Learning
Statistical Machine Learning
Optimization
AI Applications

Education

Ph.D. in Electrical Engineering and Computer Sciences, 2015 - 2022

University of California, Berkeley
B.Sc. in Electrical Engineering, 2011 - 2015

Sharif University of Technology

Publications and Patents

Baihe Huang, Paria Rashidinejad, Hanlin Zhu, Jiantao Jiao (2023). Efficient Online and Offline Reinforcement Learning via Offline Regression Oracles. Preprint.

Hanlin Zhu, Paria Rashidinejad, Jiantao Jiao (2023). Importance Weighted Actor-Critic for Optimal Conservative Offline Reinforcement Learning. Preprint.

Cite ArXiv

Paria Rashidinejad, Hanlin Zhu, Kunhe Yang, Stuart Russell, Jiantao Jiao (2022). Optimal Conservative Offline RL with General Function Approximation via Augmented Lagrangian. In International Conference on Learning Representations (ICLR) 2023 (Spotlight, top 8%).

PDF Cite ArXiv Reviews

Paria Rashidinejad, Banghua Zhu, Cong Ma, Jiantao Jiao, Stuart Russell (2021). Bridging Offline Reinforcement Learning and Imitation Learning: A Tale of Pessimism. In Advances in Neural Information Processing Systems (NeurIPS) 2021 and IEEE Transactions in Information Theory 2022..

PDF Cite Slides Video

Tianjun Zhang, Paria Rashidinejad, Jiantao Jiao, Yuandong Tian, Joseph Gonzalez, Stuart Russell (2021). MADE: Exploration via Maximizing Deviation from Explored Regions. In Advances in Neural Information Processing Systems (NeurIPS) 2021.

PDF Cite Code Video

Paria Rashidinejad, Jiantao Jiao, Stuart Russell (2020). SLIP: Learning to Predict in Unknown Dynamical Systems with Long-Term Memory. In Advances in Neural Information Processing Systems (NeurIPS) 2020 (Oral, top 1%).

PDF Cite Code Slides Video

Paria Rashidinejad, Xiao Hu, Stuart Russell (2020). Patient-Adaptable Intracranial Pressure Morphology Analysis Using a Probabilistic Model-Based Approach. In Physiological Measurement.

Cite iopscience

Paria Rashidinejad, Navaneeth Jamadagni, Arun Raghavan, Craig Schelp, Charles Gordon (2020). Techniques for Accurately Estimating the Reliability of Storage Systems. U.S. Patent No. 11,416,324..

PDF Cite