Max Qiushi Lin

I am a PhD student in Computer Science at Simon Fraser University, where I am fortunate to be advised by Sharan Vaswani. Prior to that, I worked with Hang Ma.

I am interested in machine learning theory. My current research explores the theoretical foundations of reinforcement learning and the development of principled and implementable algorithms that enable agents to learn from experience. I am also intrigued by probability theory, optimization, and statistics.

Papers

  1. Provably Efficient Reinforcement Learning with General Utilities in Bilinear MDPs

    Max Qiushi Lin, Ahmed Magd, and Sharan Vaswani

    In Submission.

  2. Augmented Lagrangian Method for Last-Iterate Convergence for Constrained MDPs

    Michael Lu, Max Qiushi Lin, Mo Chen, and Sharan Vaswani

    Preprint, 2026.

  3. Optimistic Actor-Critic with Parametric Policies for Linear Markov Decision Processes

    Max Qiushi Lin, Reza Asad, Kevin Tan, Haque Ishfaq, Csaba Szepesvári, and Sharan Vaswani

    NeurIPS Workshop on Aligning Reinforcement Learning Experimentalists and Theorists (ARLET), 2025.
    AISTATS, 2026.

  4. Rethinking the Global Convergence of Softmax Policy Gradient with Linear Function Approximation: The Case of Multi-Armed Bandits

    Max Qiushi Lin, Jincheng Mei, Matin Aghaei, Michael Lu, Bo Dai, Alekh Agarwal, Dale Schuurmans, Csaba Szepesvári, and Sharan Vaswani

    Preprint, 2025.