Marek Petrik

University of New Hampshire, University of New Hampshire at Manchester

Papers

Total Citations

H-Index

About

Marek Petrik is a leading researcher in robotics and artificial intelligence, with a primary focus on imitation learning, inverse reinforcement learning, and human-robot interaction. His work addresses the critical challenge of enabling robots to learn complex behaviors from human demonstrations, even when those demonstrations are imperfect or noisy. Petrik’s most influential paper, “Inverse Reinforcement Learning of Interaction Dynamics from Demonstrations” (2019, 8 citations), introduced a novel framework for inferring the underlying reward functions that drive sequential tasks, allowing robots to generate policies that faithfully mimic expert behavior. This contribution is foundational for advancing autonomous systems in collaborative environments. Building on this, his 2021 work, “Robust Behavior Cloning with Adversarial Demonstration Detection” (5 citations), tackles a practical limitation in imitation learning: the assumption that all demonstrations are correct. By developing methods to detect and filter out adversarial or erroneous demonstrations, Petrik significantly enhances the robustness and real-world applicability of learning-from-demonstration systems. His research bridges the gap between theoretical models and practical deployment, making him a key figure in the development of safer, more reliable autonomous robots.