Next
Livestream will start soon!
Livestream has already ended.
Presentation has not been recorded yet!
  • title: Best of Both Worlds Policy Optimization
      0:00 / 0:00
      • Report Issue
      • Settings
      • Playlists
      • Bookmarks
      • Subtitles Off
      • Playback rate
      • Quality
      • Settings
      • Debug information
      • Server sl-yoda-v2-stream-004-alpha.b-cdn.net
      • Subtitles size Medium
      • Bookmarks
      • Server
      • sl-yoda-v2-stream-004-alpha.b-cdn.net
      • sl-yoda-v2-stream-004-beta.b-cdn.net
      • 1685195716.rsc.cdn77.org
      • 1239898752.rsc.cdn77.org
      • Subtitles
      • Off
      • English
      • Playback rate
      • Quality
      • Subtitles size
      • Large
      • Medium
      • Small
      • Mode
      • Video Slideshow
      • Audio Slideshow
      • Slideshow
      • Video
      My playlists
        Bookmarks
          00:00:00
            Best of Both Worlds Policy Optimization
            • Settings
            • Sync diff
            • Quality
            • Settings
            • Server
            • Quality
            • Server

            Best of Both Worlds Policy Optimization

            Jul 25, 2023

            Speakers

            CD

            Christoph Dann

            Speaker · 0 followers

            CW

            Chen-Yu Wei

            Speaker · 0 followers

            JZ

            Julian Zimmert

            Speaker · 0 followers

            About

            Policy optimization methods are popular reinforcement learning algorithms in practice and recent works have build theoretical foundation for them by proving $\sqrt{T}$ regret bounds even when the losses are adversarial. Such bounds are tight in the worst case but often overly pessimistic. In this work, we show that by carefully designing the regularizer, bonus terms, and learning rates, one can achieve a more favorable $\text{polylog}(T)$ regret bound when the losses are stochastic, without sacr…

            Organizer

            I2
            I2

            ICML 2023

            Account · 615 followers

            Like the format? Trust SlidesLive to capture your next event!

            Professional recording and live streaming, delivered globally.

            Sharing

            Recommended Videos

            Presentations on similar topic, category or speaker

            The Regret of Exploration and the Control of Bad Episodes in Reinforcement Learning
            05:20

            The Regret of Exploration and the Control of Bad Episodes in Reinforcement Learning

            Victor Boone, …

            I2
            I2
            ICML 2023 2 years ago

            Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%

            Rethinking Warm-Starts with Predictions: Learning Predictions Close to Sets of Optimal Solutions for Faster L-/L-Convex Function Minimization
            04:37

            Rethinking Warm-Starts with Predictions: Learning Predictions Close to Sets of Optimal Solutions for Faster L-/L-Convex Function Minimization

            Shinsaku Sakaue, …

            I2
            I2
            ICML 2023 2 years ago

            Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%

            Stable Estimation of Heterogeneous Treatment Effect
            05:04

            Stable Estimation of Heterogeneous Treatment Effect

            Anpeng Wu, …

            I2
            I2
            ICML 2023 2 years ago

            Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%

            Efficient Rate Optimal Regret for Adversarial Contextual MDPs Using Online Function Approximation
            05:14

            Efficient Rate Optimal Regret for Adversarial Contextual MDPs Using Online Function Approximation

            Orin Levy, …

            I2
            I2
            ICML 2023 2 years ago

            Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%

            Neural Priority Queues for GNNs
            13:29

            Neural Priority Queues for GNNs

            Rishabh Jain, …

            I2
            I2
            ICML 2023 2 years ago

            Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%

            StriderNet: A Graph Reinforcement Learning Approach to Optimize Atomic Structures on Rough Energy Landscapes
            05:15

            StriderNet: A Graph Reinforcement Learning Approach to Optimize Atomic Structures on Rough Energy Landscapes

            Vaibhav Bihani, …

            I2
            I2
            ICML 2023 2 years ago

            Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%

            Interested in talks like this? Follow ICML 2023