Addison Kalanther

I am a researcher in Berkeley Artificial Intelligence Research at UC Berkeley, advised by Professor Shankar Sastry. I work on reinforcement learning methods for multi-agent games with unknown adversaries. I graduated with my bachelor's degree in EECS from UC Berkeley in Winter 2024. I previously worked as a software engineer, interning at Salesforce, Lucid Software, and BeamNG.

I am currently applying to PhD programs in robotics and AI. If you are interested in my research, please reach out to me.

Email  /  Scholar  /  X  /  LinkedIn  /  GitHub

Research

I am interested in robot learning for non-cooperative environments.

We train a team of heterogeneous UAVs to coordinate and guide human evacuees to safety in an urban fire scenario. We model the problem as a POMDP where humans have unknown panicked behavior and use reinforcement learning to learn adaptive policies for guiding evacuees.

We develop a sample-efficient adaptation strategy to handle unknown adversaries in a partially-observable pursuit-evasion game. We use a bounded rationality model to collect diverse and sophisticated evader policies and train an adaptive policy for the pursuer team using deep RL.

Design and source code from Jon Barron's website.