Chen Tessler

About

Chen Tessler is a researcher specializing in robust reinforcement learning and continuous control, with a particular focus on developing algorithms that remain reliable under uncertainty and adversarial conditions. His most recognized contribution, "Action Robust Reinforcement Learning and Applications in Continuous Control" (2019), has garnered 66 citations and represents a significant advancement in the field of safe and dependable AI systems. In this work, Tessler formalized novel criteria for robustness to action uncertainty, addressing scenarios where an agent's intended actions may be perturbed or corrupted — a critical challenge in deploying reinforcement learning to real-world continuous control tasks such as robotics and autonomous systems. By rigorously defining and solving for policies that maximize reward even under adversarial interference, Tessler helped bridge the gap between theoretical reinforcement learning guarantees and practical deployment reliability. His research speaks directly to growing concerns in the AI community about the brittleness of learned policies in dynamic, unpredictable environments. Tessler's work provides foundational tools for researchers and engineers seeking to build reinforcement learning agents that are not merely high-performing under ideal conditions, but genuinely resilient when facing real-world imperfections.

Research Focus

Key Achievements

Top Papers

1. 1
   Action Robust Reinforcement Learning and Applications in Continuous Control

   66 citations · 2019

Key Collaborators

Contact & Links