Daniele Reda
Email: dreda at cs dot ubc dot ca

I am a PhD student at University of British Columbia exploring applications of reinforcement learning to robotics with professor Michiel Van de Panne. I have a wide range of interests but my current research focus is mostly on representation learning applied to reinforcement learning problems and most specifically on how we can learn meaningful and interpretable representations of our surrounding from high-dimension and unstructured inputs like vision.

Before, I was a reinforcement learning researcher at Wayve, applying machine learning to robotics control to teach cars how to drive themselves.

I completed my B.Sc. at Polytechnic University of Turin and my M.Sc jointly between Polytechnic University of Turin and Telecom ParisTech - Eurecom Research Center. I was a visiting researcher at Berkeley AI Research at UC Berkeley where I did research for my master thesis related to graphical models and statistical learning with professor Ruzena Bajcsy.

Sometimes I write about random stuff over here.

CV | Google Scholar | Github | Linkedin | Twitter

Publications
Urban Driving with Conditional Imitation Learning
Jeffrey Hawke* Richard Shen*, Corina Gurau*, Siddharth Sharma*, Daniele Reda*, Nikolay Nikolov*, Przemyslaw Mazur*, Sean Micklethwaite*, Nicolas Griffiths*, Amar Shah*, Alex Kendall*
(* indicates equal contribution)
International Conference on Robotics and Automation (ICRA), 2020
(Also in ML for Autonomous Driving Workshop at NeurIPS, 2019)

arXiv | blog post | video
Learning to drive in a day
Alex Kendall, Jeffrey Hawke, David Janz, Przemyslaw Mazur, Daniele Reda, John-Mark Allen, Vinh-Dieu Lam, Alex Bewley, Amar Shah
International Conference on Robotics and Automation (ICRA), 2019
(Also in ML for Intelligent Transportation Systems Workshop at NeurIPS, 2018)

arXiv | blog post | video
Teaching

Template: this and this