Yi-Shan Wu

15 papers A* 3A 2Journal 10
YearRankTypeTitle / Venue / Authors
2025 J jnl
Deep Actor-Critics with Tight Risk Certificates.
CoRR
Bahareh Tasdighi, Manuel Haussmann, Yi-Shan Wu, Andrés R. Masegosa, Melih Kandemir
2025 A conf
Deep Exploration with PAC-Bayes.
ECAI
Bahareh Tasdighi, Manuel Haussmann, Nicklas Werge, Yi-Shan Wu, Melih Kandemir
2025 J jnl
Directional Ensemble Aggregation for Actor-Critics.
CoRR
Nicklas Werge, Yi-Shan Wu, Bahareh Tasdighi, Melih Kandemir
2025 A conf
Improving Actor-Critic Training with Steerable Action-Value Approximation Errors.
ECAI
Bahareh Tasdighi, Nicklas Werge, Yi-Shan Wu, Melih Kandemir
2025 J jnl
ObjectRL: An Object-Oriented Reinforcement Learning Codebase.
CoRR
Gulcin Baykal, Abdullah Akgül, Manuel Haussmann, Bahareh Tasdighi, Nicklas Werge, Yi-Shan Wu, Melih Kandemir
2024 J jnl
Exploring Pessimism and Optimism Dynamics in Deep Reinforcement Learning.
CoRR
Bahareh Tasdighi, Nicklas Werge, Yi-Shan Wu, Melih Kandemir
2024 J jnl
Probabilistic Actor-Critic: Learning to Explore with PAC-Bayes Uncertainty.
CoRR
Bahareh Tasdighi, Nicklas Werge, Yi-Shan Wu, Melih Kandemir
2024 A* conf
Recursive PAC-Bayes: A Frequentist Approach to Sequential Prior Updates with No Information Loss.
NeurIPS
Yi-Shan Wu, Yijie Zhang, Badr-Eddine Chérief-Abdellatif, Yevgeny Seldin
2024 J jnl
Recursive PAC-Bayes: A Frequentist Approach to Sequential Prior Updates with No Information Loss.
CoRR
Yi-Shan Wu, Yijie Zhang, Badr-Eddine Chérief-Abdellatif, Yevgeny Seldin
2024 J jnl
The Cold Posterior Effect Indicates Underfitting, and Cold Posteriors Represent a Fully Bayesian Method to Mitigate It.
Trans. Mach. Learn. Res.
Yijie Zhang, Yi-Shan Wu, Luis A. Ortega, Andrés R. Masegosa
2023 J jnl
If there is no underfitting, there is no Cold Posterior Effect.
CoRR
Yijie Zhang, Yi-Shan Wu, Luis A. Ortega, Andrés R. Masegosa
2022 A* conf
Split-kl and PAC-Bayes-split-kl Inequalities for Ternary Random Variables.
NeurIPS
Yi-Shan Wu, Yevgeny Seldin
2022 J jnl
Split-kl and PAC-Bayes-split-kl Inequalities.
CoRR
Yi-Shan Wu, Yevgeny Seldin
2021 A* conf
Chebyshev-Cantelli PAC-Bayes-Bennett Inequality for the Weighted Majority Vote.
NeurIPS
Yi-Shan Wu, Andrés R. Masegosa, Stephan Sloth Lorenzen, Christian Igel, Yevgeny Seldin
2021 J jnl
Chebyshev-Cantelli PAC-Bayes-Bennett Inequality for the Weighted Majority Vote.
CoRR
Yi-Shan Wu, Andrés R. Masegosa, Stephan Sloth Lorenzen, Christian Igel, Yevgeny Seldin