Learning from Interventions

We study how human interventions, like resets, corrections, and implicit signals, among others, can serve as a principled source of structure for safe and efficient robot learning. By interpreting interventions as defining support sets of safe and high-value states, we design algorithms that provably accelerate convergence while bounding suboptimality. This framework extends to learning intervention models directly from observation, allowing agents to autonomously infer when to reset, request help, or modify their behavior. Our broader goal is to establish a unified foundation for learning under human guidance, where safety, efficiency, and adaptability emerge naturally from the dynamics of interaction.

Published contributors

Siddhartha Srinivasa
Ethan Pronovost
Sanjiban Choudhury
Matthew Barnes
Matthew Schmittle
Samuel Ainsworth

Publications

BibTeX of these publications (.bib)

Robust Intervention Learning from Emergency Stop Interventions

Ethan Pronovost, Khimya Khetarpal, Siddhartha Srinivasa

arXiv, 2026

PDF arXiv Website interventions

Expert Intervention Learning: An online framework for robot learning from explicit and implicit human feedback

J. Spencer, Sanjiban Choudhury, M. Barnes, M. Schmittle, M. Chiang, P. Ramadge, S.S. Srinivasa

Autonomous Robots, 46, pp. 99-113, 2022

PDF interventions

Learning from interventions: Human-robot interaction as both explicit and implicit feedback

J. Spencer, Sanjiban Choudhury, M. Barnes, M. Schmittle, M. Chiang, P. Ramadge, S.S. Srinivasa

Robotics: Science and Systems, 2020

PDF interventions

Mo States Mo Problems: Emergency Stop Mechanisms from Observation

S. Ainsworth, M. Barnes, S.S. Srinivasa

Advances in Neural Information Processing Systems, 2019

PDF interventions