Amrit Krishnan

About

Amrit Krishnan is a researcher focused on the intersection of reinforcement learning (RL), robust control, and autonomous systems. His work primarily investigates the reliability and resilience of deep RL algorithms, particularly in continuous control tasks critical for robotics and autonomous navigation. In his most-cited paper, "Characterising the Robustness of Reinforcement Learning for Continuous Control using Disturbance Injection" (2022), Krishnan leverages an open-source benchmark suite to systematically inject disturbances—such as sensor noise and actuator faults—into state-of-the-art deep and robust RL algorithms. This work provides a rigorous framework for evaluating how well these algorithms withstand real-world perturbations, revealing vulnerabilities in standard approaches and offering insights for designing more dependable controllers. While his citation count is still growing, Krishnan’s contributions are notable for bridging the gap between theoretical RL and practical deployment, emphasizing the importance of benchmarking robustness in continuous action spaces. His research is particularly valuable for students and engineers seeking to build safer, more resilient autonomous systems, and his methodology has laid groundwork for future studies on fault-tolerant learning in dynamic environments.

Research Focus

Key Achievements

Top Papers

1. 1
   Characterising the Robustness of Reinforcement Learning for Continuous Control using Disturbance Injection

   2 citations · 2022

Key Collaborators

Contact & Links