Sotabase

Career

· Postdoctoral Researcher in Computational Neuroscience, McGill University2025–
· Research Intern, Google DeepMind2021–2023
· PhD in Computer Science (AI/RL/neuro), McGill University2019–2025
· Machine Learning Engineer, Apsisware / Sparktech Software2016–2018
· M.Sc in Computer Science (AI), University Politehnica of Bucharest2016–2018
· Software Engineer, Deutsche Bank / Misys / Cronian Labs2013–2015
· B.Sc in Computer Science (math/CS), University Politehnica of Bucharest2009–2013

Publications (16)

Forethought and Hindsight in Credit Assignment
Neural Information Processing Systems · 2020
27
cited
Learning Expected Emphatic Traces for Deep RL
AAAI Conference on Artificial Intelligence · 2021
12
cited
Selective Credit Assignment
arXiv.org · 2022
3
cited
Functional Acceleration for Policy Mirror Descent
arXiv.org · 2024
2
cited
IFT 6085 - Lecture 5 Accelerated Methods - Polyak’s Momentum (Heavy Ball Method)
2019
2
cited
A Generalized Bootstrap Target for Value-Learning, Efficiently Combining Value and Feature Predictions
AAAI Conference on Artificial Intelligence · 2022
1
cited
Acceleration in Policy Optimization
2023
Excitatory-Inhibitory Dynamics in Adaptive Decision-Making
International Conference on Neural Information Processing · 2025
Optimism and Adaptivity in Policy Optimization
arXiv.org · 2023
Recurrent Policies Are Not Enough for Continual Reinforcement Learning
Sotabase
Veronica Chelu | Researcher Profile | Sotabase | Sotabase