Learning Shared Safety Constraints from Multi-task Demonstrations
Jul 28, 2023
Speakers
Gokul Swamy
Speaker · 0 followers
Konwoo Kim
Speaker · 0 followers
Sanjiban Choudhury
Speaker · 0 followers
About
Regardless of the particular task we want them to perform in an environment, there are often shared *safety constraints* we want our agents to respect. For example, regardless of whether it is making a sandwich or clearing the table, a kitchen robot should not break a plate. Manually specifying such a constraint can be both time-consuming and error-prone. We show how to learn constraints from expert demonstrations of safe task completion by extending inverse reinforcement learning (IRL) techniques to the space of constraints. Intuitively, we learn constraints that forbid highly rewarding behavior that the expert could have taken but chose not to. Unfortunately, the constraint learning problem is rather ill-posed and typically leads to overly conservative constraints that forbid all behavior that the expert did not take. We counter this by leveraging diverse demonstrations that naturally occur in multi-task settings to learn a tighter set of constraints. We validate our method with simulation experiments on high-dimensional continuous control tasks.Regardless of the particular task we want them to perform in an environment, there are often shared *safety constraints* we want our agents to respect. For example, regardless of whether it is making a sandwich or clearing the table, a kitchen robot should not break a plate. Manually specifying such a constraint can be both time-consuming and error-prone. We show how to learn constraints from expert demonstrations of safe task completion by extending inverse reinforcement learning (IRL) techniqu…
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
On the Role of Attention in Prompt-tuning
Watch later
Samet Oymak, …
A Theoretical Analysis of the Learning Dynamics under Class Imbalance
Watch later
Training Spiking Neural Networks with Local Tandem Learning
Watch later
Toward Fast Grad-Based Value Estimation
Watch later
Beyond Reward: Offline Preference-guided Policy Optimization
Watch later
Yachen Kang, …
