Counterexample Guided RL Policy Refinement Using Bayesian Optimization
6. prosince 2021
Řečníci
Briti Gangopadhyay
Řečník · 0 sledujících
Pallab Dasgupta
Řečník · 0 sledujících
O prezentaci
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…
Organizátor
NeurIPS 2021
Účet · 1,9k sledujících
O organizátorovi (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.
Baví vás formát? Nechte SlidesLive zachytit svou akci!
Profesionální natáčení a streamování po celém světě.
Sdílení
Doporučená videa
Prezentace na podobné téma, kategorii nebo přednášejícího
Policy Optimization via Optimal Policy Evaluation
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Learning to Iteratively Solve Routing Problems with Dual-Aspect Collaborative Transformer
Zhlédnout později
Yining Ma, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Smooth Normalizing Flows
Zhlédnout později
Jonas Köhler, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Video Instance Segmentation using Inter-Frame Communication Transformers
Zhlédnout později
Sukjun Hwang, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
An Image is Worth More Than a Thousand Words: Towards Disentanglement in The Wild
Zhlédnout později
Aviv Gabbay, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Learning Semantic Representations to Verify Hardware Designs
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %