Bogdan Mazoure

Bogdan Mazoure

Research Scientist at Apple MLR.

Bogdan Mazoure - Research Scientist at Apple MLR

I am taking interns, as well as students to be co-advised with faculty members.

News

  • CVL is accepted to CoRL 2023 as a poster [Paper].
  • Accelerating exploration and representation learning with offline pre-training is accepted to ICML 2023 as a workshop poster [Paper].
  • Learning About Progress From Experts is accepted to ICLR 2023 as a spotlight (top 25%) [Paper].
  • I started at the MLR team at Apple as a Research Scientist.
  • CVL is accepted to NeurIPS 2022 Deep RL workshop [Paper].
  • GSF is accepted to NeurIPS 2022 as a poster [Paper].
  • Low-Rank Representation of Reinforcement Learning Policies is accepted to JAIR [Paper].
  • Sequential Density Estimation via Nonlinear Continuous Weighted Finite Automata is accepted to LearnAut 2022 [Paper].
  • Short-Horizon Policy Iteration (jointly with Microsoft Research) is accepted to ECML-PKDD 2022. [Paper].
  • Our generalization quantification toolbox (jointly with Microsoft Research) is out. [Link].
  • CTRL is accepted to ICLR 2022 as poster. [Paper].
  • I am teaching the COMP 424: Artificial Intelligence class at McGill University during the Winter 2022 term. [Link].
  • GSF is accepted to NeurIPS 2021 Offline RL workshop as poster. [Paper].
  • I am teaching the BINF 7105: Méthodes statistiques en bioinformatique class at UQAM University during the Fall 2021 term. [Link].
  • Our COVID-19 phylogenetic analysis is accepted to BMC Ecology and Evolution. [Paper].
  • NTK-CL is accepted to AISTATS 2021 as a poster. [Paper][Talk].
  • DRIML is accepted to NeurIPS 2020 as poster. [Paper] [Blog] [Poster].
  • UQF is accepted to AISTATS 2020 as a poster. [Paper][Poster].

About me

I am currently a Research Scientist at the Machine Learning Research team at Apple, working with Josh Susskind, Walter Talbott and Devon Hjelm on fundamental problems of representation learning for sequential decision making tasks.

I recently defended my PhD at the Montreal Institute for Learning Algorithms (MILA) and McGill University, co-supervised by Devon Hjelm and Doina Precup. My research interests include deep reinforcement learning, probabilistic modeling, variational inference and representation learning.

I also was a research intern at DeepMind, working with Ankit Anand, Jake Bruce and Rob Fergus on unsupervised pre-training of state representations for efficient RL finetuning.

In the summer of 2021, I was interning in the Robotics team at Google Brain, with Jonathan Tompson and Ofir Nachum, working on using self-supervised learning to improve generalization capabilities of offline RL agents. I was a research intern at Microsoft Research, New York in the reinforcement learning team during summer 2020, working on counterfactual evaluation in contextual bandits. Previously, I was a research intern at Microsoft Research Montreal in the reinforcement learning team during summer 2019. I was also a research intern at Nuance during the summer of 2018 where I collaborated with Atta Norouzian. My work there focused on modeling acoustic signals such as speech with deep neural architectures.

I have completed my Master’s in Statistics at McGill University under the supervision of Johanna Neslehova. My thesis focused on reconstructing graphical models from discrete data with variational inference and multiarmed bandits. It can be found here: link. Before that, I obtained a Bachelor’s in Computer Science and Statistics in 2017 from McGill University.

Research interests

  • Deep reinforcement and representation learning;

  • High-dimensional statistics and optimization;

  • Parametric and non-parametric Bayesian methods, approximate inference;

  • Probabilistic graphical models;

  • Generative models and density estimation.

Work

  • Student Researcher (Now) Google Brain
  • Research intern (Summer 2021) Google Brain
  • Researcher (2020-2021) Microsoft Research
  • Research intern (Summer 2020) Microsoft Research (NYC)
  • Research intern (Summer 2019) Microsoft Research (Montreal)
  • Research intern (Summer 2018) Nuance

Education

🎓 PhD in Computer Science (2019-) McGill University / MILA

🎓 MSc in Math & Stats (2017-2018) McGill University

🎓 BSc in Stats & Computer Science (2014-2017) McGill University