Fabio Pardo

Fabio Pardo


PhD Student in Machine Learning

Imperial College London
London, SW7 1NA
United Kingdom

f.pardo[at]imperial.ac.uk

fabiopardo @pardofab

I am a fourth-year PhD student in the Robot Intelligence Lab at Imperial College London.
My main research focuses on Deep Reinforcement Learning.

Education

2016 – present
PhD in Machine Learning
Deep Reinforcement Learning
@ Imperial College, London, UK
2014 – 2015
Master's degree in Computer Science
AI, ML, Robotics
@ Pierre et Marie Curie University, Paris, France
2012 – 2014
Master’s degree in Cognitive Science
Neuroscience, Cognitive Psychology, Computational Modeling, Neuroimaging, AI
@ École Normale Supérieure Ulm, EHESS and Descartes University, Paris, France
2009 – 2012
Bachelor's degree in Computer Science
@ Pierre et Marie Curie University, Paris, France

Research Internships

2019
Motor primitives and competitive self-play
Raia Hadsell, Nicolas Heess, Josh Merel and Leonard Hasenclever
@ DeepMind, London, UK
2015
Deep reinforcement learning for autonomous robot navigation from vision
Tetsunari Inamura
@ National Institute of Informatics, Tokyo, Japan
2014
Multimodal concepts emergence for a humanoid robot in interaction with a human tutor
David Filliat
@ Flowers laboratory, Inria and ENSTA ParisTech, Paris, France
Video and video
2013
Optimal decision making based on a mixture of prediction experts
Homeostatic engine for reinforcement learning agents
Laurent Orseau
@ Inria and AgroParisTech, Paris, France
2013
Ontology visualization methods and their impact on short-term memory storage in humans
Jean-Gabriel Ganascia
@ Lip6, Paris, France

Publications

CoMic: Complementary Task Learning & Mimicry for Reusable Skills
Leonard Hasenclever, Fabio Pardo, Raia Hadsell, Nicolas Heess, Josh Merel
@ ICML 2020
Paper and code
Scaling All-Goals Updates in Reinforcement Learning Using Convolutional Neural Networks
Fabio Pardo, Vitaly Levdik, Petar Kormushev
@ AAAI 2020
@ NeurIPS 2018 Deep RL Workshop
@ ICML 2018 Exploration in RL Workshop
Paper, poster, website and code
Time Limits in Reinforcement Learning
Fabio Pardo, Arash Tavakoli, Vitaly Levdik, Petar Kormushev
@ ICML 2018
@ NIPS 2017 Deep RL Symposium
Paper, poster and website
Action Branching Architectures for Deep Reinforcement Learning
Arash Tavakoli, Fabio Pardo, Petar Kormushev
@ AAAI 2018
@ NIPS 2017 Deep RL Symposium
Paper and poster

Softwares



Tonic: A Deep Reinforcement Learning Library for Fast Prototyping and Benchmarking
  • a collection of configurable modules such as: exploration strategies, memories, neural networks, and optimizers
  • a collection of baseline agents built with these modules
  • support for the two most popular deep learning frameworks: TensorFlow and PyTorch
  • support for the three most popular sets of continuous control environments: OpenAI Gym, DeepMind Control Suite and PyBullet
  • a large-scale benchmark of the baseline agents on 70 tasks
  • scripts to train in a reproducible way, plot results, and play with trained agents.
Repository and data

Miscellaneous

2018 – present
Organizer of Imperial's deep reinforcement learning reading group
Weekly meetings to discuss papers (contact me if you want to join)
@ Imperial College, London, UK
2016 – 2018
Graduate teaching assistant in Computing and Robotics
Lectures, tutorials, exams
@ Imperial College, London, UK
2011 and 2012
Twice finalist of Prologin, the French national programming contest
Algorithmic tests and 36-hour hackathon
@ École Polytechnique and EPITA, Paris, France