PUBLICATIONS

Preprints

• Exploring Pessimism and Optimism Dynamics in Deep Reinforcement Learning [paper]
  Bahareh Tasdighi, Nicklas Werge, Yi-Shan Wu, Melih Kandemir
  preprint. 2024

• Probabilistic Actor-Critic: Learning to Explore with PAC-Bayes Uncertainty [paper]
  Bahareh Tasdighi, Nicklas Werge, Yi-Shan Wu, Melih Kandemir

Journal Papers

• The Cold Posterior Effect Indicates Underfitting, and Cold Posteriors Represent a Fully Bayesian Method to Mitigate It [paper]
  Yijie Zhang, Yi-Shan Wu, Luis A. Ortega, Andrés R. Masegosa
  TMLR 2024

Conference Papers

• Recursive PAC-Bayes: A Frequentist Approach to Sequential Prior Updates with No Information Loss [paper]
  Yi-Shan Wu, Yijie Zhang, Badr-Eddine Chérief-Abdellatif, Yevgeny Seldin
  In Advances in Neural Information Processing Systems (NeurIPS), 2024 (spotlight)

• Split-kl and PAC-Bayes-split-kl Inequalities For Ternary Random Variables [paper]
  Yi-Shan Wu, Yevgeny Seldin
  In Advances in Neural Information Processing Systems (NeurIPS), 2022.

• Chebyshev-Cantelli PAC-Bayes-Bennett Inequality for the Weighted Majority Vote [paper]
  Yi-Shan Wu, Andres Masegosa, Stephan Lorenzen, Christian Igel, Yevgeny Seldin
  In Advances in Neural Information Processing Systems (NeurIPS), 2021.

• Lifelong Learning with Branching Experts [paper]
  Yi-Shan Wu, Yi-Te Hong and Chi-Jen Lu
  In Proceedings of the Asian Conference on Machine Learning (ACML), 2021.

• Lifelong Optimization with Low Regret  [paper]
  Yi-Shan Wu, Po-An Wang and Chi-Jen Lu
  In Proceedings of the International Conference on Artificial Intelligence and Statistics (AISTATS), 2019.