Brendan Crowe

University of New Hampshire at Manchester

Papers

Total Citations

H-Index

About

Brendan Crowe is a rising researcher in robotics and machine learning, whose work focuses on making imitation learning (IL) more robust and practical for real-world deployment. His key contributions address a critical flaw in traditional IL frameworks: the assumption that all expert demonstrations are correct. In his highly cited 2021 paper, "Robust Behavior Cloning with Adversarial Demonstration Detection" (5 citations), Crowe proposed a novel method that detects and filters out adversarial or erroneous demonstrations, allowing robots to learn effectively even from imperfect human input. This work has significant implications for safety-critical applications, from autonomous driving to assistive robotics, where noisy or malicious data can lead to catastrophic failures. By challenging the idealized assumptions of IL, Crowe is helping to bridge the gap between controlled lab environments and the unpredictable real world. His research is essential reading for students and engineers seeking to build more resilient, trustworthy learning systems that can operate reliably alongside humans.