Vincent Luczkow

Vincent Luczkow is a machine learning researcher and engineer based in the San Francisco Bay Area. He completed his MSc at Mila (Montreal Institute for Learning Algorithms) in 2020, and subsequently pursued a PhD at McGill University under the supervision of Doina Precup and Prakash Panangaden, with a research focus on reinforcement learning and causality. His thesis explored structural causal models for reinforcement learning, introducing approaches to learn minimal causal models of environments at the time-step level. In 2020, he received a $10,000 grant from the Long-Term Future Fund to support research on counterfactual impact minimization, which connects to AI safety work on designing agents that avoid unintended side effects. He has since worked at insitro, a machine learning drug discovery company. He is also active in the forecasting community through Manifold Markets and maintains a personal website and GitHub under the handle vluzko.

No endorsements yet.

Personal Website

https://vluzko.github.io/

LinkedIn

LinkedIn

Twitter / X

Twitter

LessWrong

-

EA Forum

-