Summary
Arash Tavakoli is a Staff Research Scientist with 11 years of experience specializing in reinforcement learning, deep learning, and multiagent systems, currently based in Los Angeles and working at Riot Games. His research advances practical RL for the "big world," with contributions on structural and temporal credit assignment, model learning under aleatoric uncertainty, and high-dimensional/multi-action problems. He combines a strong academic pedigree—a PhD in Machine Learning from Imperial College and postdoc work at the Max Planck Institute and visits with Sutton—with industry research stints at Microsoft and Shift Lab, bridging theory and real-world applications. Arash is particularly focused on model-based RL and unsupervised skill discovery for long-horizon, non-stationary tasks and is exploring applied deployments in robotics, recommender systems, and financial markets. A less obvious strength is his track record of navigating both foundational theory and engineering constraints, making his work well-suited to productionizing complex RL systems.
11 years of coding experience
8 years of employment as a software developer
Master of Science (MSc), Computer Science, Master of Science (MSc), Computer Science at University of Southern California
Exchange Student, Electrical and Computer Engineering, Exchange Student, Electrical and Computer Engineering at Georgia Institute of Technology
Doctor of Philosophy (PhD), Machine Learning, Doctor of Philosophy (PhD), Machine Learning at Imperial College London
Master of Engineering (MEng), Electrical Engineering, First-Class Honours, Master of Engineering (MEng), Electrical Engineering, First-Class Honours at University College London, U. of London