Addison Kalanther

I’m an incoming EECS PhD Student at UC Berkeley. I work in Berkeley Artificial Intelligence Research (BAIR), advised by Professor Shankar Sastry. I am interested in general intelligence and recursive self improvement for robotics, and my research concerns robot learning, reinforcement learning, and agentic systems. More broadly, I am interested in the makeup of intelligence. Outside of research, I enjoy sports and music.

I’m also a researcher at a stealth startup and an editor at BAIR blog.

Research

NePPO: Near-Potential Policy Optimization for General-Sum Multi-Agent Reinforcement Learning

A. Kalanther, S. Bharvirkar, S. Sastry, C. Maheshwari
IEEE Conference on Decision and Control (CDC), 2026 [Accepted]
arXiv
We propose NePPO, a multi-agent RL algorithm for approximating Nash equilibria in general-sum games. It learns a player-independent potential function by minimizing a novel objective with zeroth-order gradient descent.

Coordinated Autonomous Drones for Human-Centered Fire Evacuation in Partially Observable Urban Environments

M. Mendoza, A. Kalanther, D. Bostwick, E. Stephan, C. Maheshwari, S. Sastry
IEEE Global Humanitarian Technology Conference (GHTC), 2025
project page / arXiv
We train heterogeneous UAVs to guide human evacuees through a simulated urban fire, modeling it as a POMDP and using RL to learn policies robust to unknown panic behavior.

Evader-Agnostic Team-Based Pursuit Strategies in Partially-Observable Environments

A. Kalanther, D. Bostwick, C. Maheshwari, S. Sastry
Workshop on Multi-Agent Systems, Robotics: Science and Systems (RSS), 2025
project page / arXiv
We develop a sample-efficient strategy for unknown adversaries in a partially-observable pursuit-evasion game, using a bounded-rationality model to collect diverse evader policies and deep RL to train an adaptive pursuer team.