Counterexample Guided RL Policy Refinement Using Bayesian Optimization
Dez 6, 2021
Sprecher:innen
Briti Gangopadhyay
Řečník · 0 sledujících
Pallab Dasgupta
Řečník · 0 sledujících
Über
Constructing Reinforcement Learning (RL) policies that adhere to safety requirements is an emerging field of study. RL agents learn via trial and error with an objective to optimize a reward signal. Often policies that are designed to accumulate rewards do not satisfy safety specifications. We present a methodology for counterexample guided refinement of a trained RL policy against a given safety specification. Our approach has two main components. The first component is an approach to discover failure trajectories using Bayesian optimization over multiple parameters of uncertainty from a policy learnt in a model-free setting. The second component selectively modifies the failure points of the policy using gradient-based updates. The approach has been tested on several RL environments, and we demonstrate that the policy can be made to respect the safety specifications through such targeted changes.Constructing Reinforcement Learning (RL) policies that adhere to safety requirements is an emerging field of study. RL agents learn via trial and error with an objective to optimize a reward signal. Often policies that are designed to accumulate rewards do not satisfy safety specifications. We present a methodology for counterexample guided refinement of a trained RL policy against a given safety specification. Our approach has two main components. The first component is an approach to discover…
Organisator
NeurIPS 2021
Účet · 1,9k sledujících
Über NeurIPS 2021
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Gefällt euch das Format? Vertraut auf SlidesLive, um euer nächstes Event festzuhalten!
Professionelle Aufzeichnung und Livestreaming – weltweit.
Freigeben
Empfohlene Videos
Präsentationen, deren Thema, Kategorie oder Sprecher:in ähnlich sind
Panel: Labor
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Oral Session 2: Reinforcement Learning
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Opening of the ImageNet: Past, Present, and Future
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Who Is the Strongest Enemy? Towards Optimal and Efficient Evasion Attacks in Deep RL
Zhlédnout později
Yanchao Sun, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Generalisation in Lifelong RL via Logical Composition
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Why Do Better Loss Functions Lead to Less Transferable Features?
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %