Course Lecturers

Lecturers

Each Lecturer will hold two/three  lessons on a specific topic. The Lecturers below are confirmed.

Maria K. Eckstein
 

Topics

Reinforcement Learning, Structure Learning, Cognitive Development, Neural Networks

Biography

Dr. Eckstein is research scientist at Google DeepMind with a PhD in psychology from UC Berkeley.

Situated at the intersection between these two fields, her work combines methods from artificial intelligence (e.g., reinforcement learning, neural networks) with those from cognitive psychology and neuroscience (e.g., controlled lab experiments, moment-to-moment neural recordings).  She is particularly interested in questions that are fundamental to both fields, including learning, decision making, cognitive representations, and structured thought.

Lectures



Auke Jan Ijspeert
 

Topics

Biorobotics, Robotics, Computational Neuroscience, Motor Control, Locomotion

Biography

Auke Ijspeert is a full professor at the EPFL (the Swiss Federal Institute of Technology at Lausanne), and head of the Biorobotics Laboratory (BioRob). He has a B.Sc./M.Sc. in physics from the EPFL (1995), and a PhD in artificial intelligence from the University of Edinburgh (1999). He carried out postdocs at IDSIA and EPFL, and at the University of Southern California (USC). He then became a research assistant professor at USC, and an external collaborator at ATR (Advanced Telecommunications Research institute) in Japan. In 2002, he came back to the EPFL as an SNF assistant professor. He was promoted to associate professor in October 2009 and to full professor in April 2016. His primary affiliation is with the Institute of Bioengineering, and secondary affiliation with the Institute of Mechanical Engineering.

His research interests are at the intersection between robotics, computational neuroscience, nonlinear dynamical systems, and applied machine learning. He is interested in using numerical simulations and robots to get a better understanding of animal locomotion and movement control, and in using inspiration from biology to design novel types of robots and locomotion controllers (see for instance Ijspeert et al Science 2007, Ijspeert Science 2014, and Nyakatura et al Nature 2019). He is also investigating how to assist people with limited mobility using exoskeletons and assistive furniture.

He is regularly invited to give talks on these topics (e.g. TED talk given at TED Global Geneva, Dec 8 2015). With his colleagues, he has received paper awards at ICRA2002, CLAWAR2005, IEEE Humanoids 2007, IEEE ROMAN 2014, CLAWAR 2015, SAB2018, and CLAWAR 2019.

He is an IEEE Fellow, member of the Board of Reviewing Editors of Science magazine, and associate editor for the IEEE Transactions on Medical Robotics and Bionics and for the International Journal of Humanoid Robotics. He has acted as an associate editor for the IEEE Transactions on Robotics (2009-2013) and for Soft Robotics (2018-2021). He was a guest editor for the Proceedings of IEEE,  IEEE Transactions on Biomedical Engineering, Autonomous Robots, IEEE Robotics and Automation Magazine, and Biological Cybernetics. He has been the organizer of 7 international conferences (BioADIT2004, SAB2004, AMAM2005, BioADIT2006, LATSIS2006, SSRR2016, AMAM2019), and a program committee member of over 50 conferences. Please visit the BioRob Home and BioRob publicationpages for more information about his research and publications (See also Ijspeert’s Google Scholar Profile).

Lectures



Zeb Kurth-Nelson
 

Topics

Neuroscience, AI

Biography

  • Honorary Professor University College London, Imaging Neuroscience
  • BS in Computer Science, 2003, Iowa State University
  • PhD in Neuroscience, 2009, University of Minnesota

Research interests

  • Deep RL models for decision making in the brain
  • Exploration, experimentation, active learning
  • Spontaneous sequences and learning and using relational maps
  • Brain-inspired network architectures

Selected publications

Kurth-Nelson, Z., Economides, M., Dolan, R. J., & Dayan, P. (2016). Fast sequences of non-spatial state representations in humans. Neuron, 91(1), 194–204. doi:https://doi.org/10.1016/j.neuron.2016.05.028

Wang, J. X., Kurth-Nelson, Z., Kumaran, D., Tirumala, D., Soyer, H., Leibo, J. Z., Hassabis, D., & Botvinick, M. (2018). Prefrontal cortex as a meta-reinforcement learning system. Nature Neuroscience, 21, 860–868.doi:https://doi.org/10.1038/s41593-018-0147-8

Dasgupta, I., Wang, J., Chiappa, S., Mitrovic, J., Ortega, P., Raposo, D., Hughes, E., Battaglia, P., Botvinick, M., & Kurth-Nelson, Z. (2019). Causal reasoning from meta-reinforcement learning. arXiv, 1901.08162. https://arxiv.org/abs/1901.08162

Dabney, W., Kurth-Nelson, Z., Uchida, N., Starkweather, C. K., Hassabis, D., Munos, R., & Botvinick, M. (2020). A distributional code for value in dopamine-based reinforcement learning. Nature Neuroscience, 577, 671–675.doi:https://doi.org/10.1038/s41586-019-1924-6

Lectures



Loic Matthey
 

Topics

Artificial General Intelligence, Machine Learning, AI, Computational Neuroscience, 

Biography

Dr. Matthey is an an ex-Neuroscientist now working on Artificial General Intelligence.

He is working as a Staff Research Scientist at Google DeepMind, more precisely on Concepts understanding, structured representation learning and RL.

Basically worked a long while on unsupervised structure learning and generative models, leveraging them to make predictions and trying to make model-based Deep Reinforcement learning work like it should.

He is mostly working with assessing massive visual-language models for their common-sense abilities, like an ex-neuroscientist would.

Lectures



Kevin J. Miller

Topics

Computational Cognitive Neuroscience

Biography

Dr. Miller is a research scientist in the DeepMind Neuroscience Lab, as well as a research fellow at University College London in the CortexLab.  He asks questions like “Can we understand what the brain is doing well enough to reproduce it in software?

His work involves training rodents to play structured games, then carefully studying both the choices that they make and the neural signals that underlie those choices. It also involves designing software agents that play the same games as the rodents, and using these agents as tools for understanding the brain.  He is excited about both reinforcement learning and neural networks as tools for building software agents, and about multi-level modeling as a tool for understanding experimental data.

Before moving to London, he did a PhD at the Princeton Neuroscience Institute with Carlos Brody and Matt Botvinick. His PhD work focused on understanding how the brain makes plans, with a focus on the orbitofrontal cortex and the hippocampus.

Lectures



Thomas Parr
 

Topics

Active Inference, Message Passing, Computational Neuroscience, Neuroscience

Biography

Thomas Parr is a theoretical neurobiologist and practising clinician. He completed his undergraduate medical studies and PhD at University College London, where he worked with Professor Karl Friston at the Wellcome Centre for Human Neuroimaging in Queen Square. He is the author of the first comprehensive textbook on Active Inference—an approach to understanding brain and behaviour from first principles. He is interested in how our brains model our environments, and how these models become dysfunctional in neurological disease. He currently works as an NIHR Academic Clinical Fellow in Neurology at the Nuffield Department of Clinical Neurosciences, University of Oxford.

Lectures



Topics

Brain-inspired computing, Neuromorphic VLSI Design, Spiking Neural Networks


Biography

Present and Previous Positions

  • 2006 – present Full professor at SISSA in the Cognitive Neuroscience Sector, from 2012 within the Area of Neuroscience
  • 2016 – 2018 Coordinator of the Area of Neuroscience at SISSA
  • 2013 – 2015 Director of the Master in Complex Actions (course in high-tech entrepreneurship) at SISSA
  • 2011 – 2013 On leave to serve as Counsellor for Scientific Affairs at the Embassy of Italy, Tel Aviv
  • 2003 – present Long-term visiting professor at NTNU, Trondheim, currently with the Centre for Neural Computation
  • 2005 – 2007 Elected board member, NENS (School Committee of the Federation of European Neuroscience Societies)
  • 2000 – 2006 Associate professor at SISSA in the Cognitive Neuroscience Sector
  • 2002 – 2006 (and again 2014-2018) Elected councilor, Jewish Community of Trieste
  • 1992 – 2000 Research fellow at SISSA, first in the Biophysics then Cognitive Neuroscience Sector
  • 1989 – 1992 Postdoc at the University of Oxford, in the Dept of Expl Psychology, laboratory of Prof ET Rolls

Research Areas

  • Hippocampus
  • Phase transitions to the faculty of language
  • Attractor neural networks
  • Evolution of the nervous systems
  • Spatial cognition
  • Statistical physics approaches to brain function

Honours and Awards

  • 1996 – present proud to have mentored PhD students who have then flourished elsewhere, including
  • Stefano Panzeri (PhD 1996, IIT Genova and Trento)
  • Francesco P Battaglia (1998, Radboud Nijmegen)
  • Yasser Roudi (2005, NTNU-Kavli Center Trondheim; Kandel Prize 2015)
  • Emilio Kropff (2007, Leloir Inst Buenos Aires; ICTP Prize 2017)
  • Athena Akrami (2010, UCL)
  • Chol Jun Kang (2017, Kim Il-Sung Univ Pyongyang) and, remotely, Mohammad Herzallah, leader of the Palestinian Neuroscience Initiative
  • 2018 Only invited neuroscientist at the Kim Il Sung Univ Intl Confer on Devel Science & Human Welfare, Pyongyang
  • 2010 Ideator, convenor (and co-organizer with O Gűntűrkűn and A Sadoyan) the Ararat Memory meeting, Yerevan
  • 2008 Elected to the Royal Norwegian Society of Sciences and Letters, Trondheim

Lectures





Tutorial Speakers

(TBA)