-
Constrained episodic reinforcement learning in concave-convex and knapsack settings
We propose an algorithm for tabular episodic reinforcement learning with...
read it
-
SafeLife 1.0: Exploring Side Effects in Complex Environments
We present SafeLife, a publicly available reinforcement learning environ...
read it
-
An empirical investigation of the challenges of real-world reinforcement learning
Reinforcement learning (RL) has proven its worth in a series of artifici...
read it
-
PAC Reinforcement Learning without Real-World Feedback
This work studies reinforcement learning in the Sim-to-Real setting, in ...
read it
-
Self-Supervised Policy Adaptation during Deployment
In most real world scenarios, a policy trained by reinforcement learning...
read it
-
Value constrained model-free continuous control
The naive application of Reinforcement Learning algorithms to continuous...
read it
-
Batch Policy Learning under Constraints
When learning policies for real-world domains, two important questions a...
read it
Robust Constrained Reinforcement Learning for Continuous Control with Model Misspecification
Many real-world physical control systems are required to satisfy constraints upon deployment. Furthermore, real-world systems are often subject to effects such as non-stationarity, wear-and-tear, uncalibrated sensors and so on. Such effects effectively perturb the system dynamics and can cause a policy trained successfully in one domain to perform poorly when deployed to a perturbed version of the same domain. This can affect a policy's ability to maximize future rewards as well as the extent to which it satisfies constraints. We refer to this as constrained model misspecification. We present an algorithm with theoretical guarantees that mitigates this form of misspecification, and showcase its performance in multiple Mujoco tasks from the Real World Reinforcement Learning (RWRL) suite.
READ FULL TEXT
Comments
There are no comments yet.