I am generally interested in unfolding consequences of design choices in machine learning pipelines (e.g. model assumptions, inductive biases, inference, optimization and model explanations) for social-technical decision making systems. I am also broadly interested in issues of AI safety, ethics and regulation. My interests include uncertainty quantification, deep generative models, deep Bayesian models, approximate inference, user modeling in RL, explanable AI and HCI.

Recent Papers

More publications can be found at my Google Scholar.

  1. Luke Bailey, Gustaf Ahdritz, Anat Kleiman, Siddharth Swaroop, Finale Doshi-Velez, Weiwei Pan, Soft prompting might be a bug, not a feature, ICML Workshop on Challenges of Deploying Generative AI Workshop, 2023.
  2. Charumathi Badrinath, Weiwei Pan, Finale Doshi-Velez, SAP-sLDA: An Interpretable Interface for Exploring Unstructured Text, ICML Workshop on Artificial Intelligence & Human-Computer Interaction, 2023.
  3. Lars L. Ankile, Brian S. Ham, Kevin Mao, Eura Shin, Siddharth Swaroop, Finale Doshi-Velez, Weiwei Pan, Discovering User Types: Characterization of User Traits by Task-Specific Behaviors in Reinforcement Learning, ICML Workshop on Artificial Intelligence & Human-Computer Interaction, 2023.
  4. Sarah Rathnam, Sonali Parbhoo, Weiwei Pan, Susan Murphy, Finale Doshi-Velez, The Unintended Consequences of Discount Regularization: Improving Regularization in Certainty Equivalence Reinforcement Learning, 40th International Conference on Machine Learning, 2023.
  5. Ziyan Zhu, Marios Mattheakis, Weiwei Pan, Efthimios Kaxiras, HubbardNet: Efficient predictions of the Bose-Hubbard model spectrum with deep neural networks, Physics Review Research, 2023.
  6. Eura Shin, Siddharth Swaroop, Weiwei Pan, Susan Murphy, Finale Doshi-Velez, Modeling Mobile Health Users as Reinforcement Learning Agents, AAAI Workshop on AI For Behavior Change, 2023.
  7. Zixi Chen, Varshini Subhash, Marton Havasi, Weiwei Pan, Finale Doshi-Velez, What Makes a Good Explanation?: A Harmonized View of Properties of Explanations, Neurips Workshop on Progress and Challenges in Building Trustworthy Embodied AI, 2022.
  8. Jiayu Yao, Yaniv Yacoby, Beau Coker, Weiwei Pan, Finale Doshi-Velez, An Empirical Analysis of the Advantages of Finite vs. Infinite Width Bayesian Neural Networks, Neurips Workshop on Gaussian Processes, Spatiotemporal Modeling, and Decision-making Systems, 2022.
  9. Yaniv Yacoby, Weiwei Pan, Finale Doshi-Velez, Mitigating the Effects of Non-Identifiability on Inference for Bayesian Neural Networks with Latent Variables, JMLR, 2022.
  10. Mark Penrod, Harrison Termotto, Varshini Reddy, Jiayu Yao, Finale Doshi-Velez, Weiwei Pan, Success of Uncertainty-Aware Deep Models Depends on Data Manifold Geometry, ICML Workshop on Responsible Decision Making in Dynamic Environments, 2022.
  11. Xin Zheng, Jiayu Yao, Finale Doshi-Velez, Weiwei Pan, From Soft Trees to Hard Trees: Gains and Losses, ICML Workshop on Responsible Decision Making in Dynamic Environments, 2022.
  12. Beau Coker, Wessel P. Bruinsma, David R. Burt, Weiwei Pan, Finale Doshi-Velez, Wide Mean-Field Variational Bayesian Neural Networks Ignore the Data, Artificial Intelligence and Statistics (AIStats), 2022.