About
Continuous advances in machine learning have enabled groundbreaking progress in diverse fields such as computer vision or strategy games by making use of artificial neural networks. In parallel, novel technologies for recording and manipulating biological neural networks allow us to probe the behavioural function of neuronal activity in unprecedented detail. By bringing together leading experts in the fields of biology, applied mathematics and physics, this symposium aims to bridge our current understanding of how biological and artificial neural networks operate.
Program
[gview file=”https://research.pasteur.fr/wp-content/uploads/2018/07/research_pasteur-neural-networks-from-brains-to-machines-and-vice-versa-booklet-v3.pdf”]
Registration
Registration is closed. There is no on-site registration.
You will need to have registered via EventBrite and replied to our follow-up email to attend the meeting.
You will not be able to access the Pasteur Campus without registration.
Sponsors
Schedule
Thursday 11th October
Physics of biological neural networks (Chair: JB Masson)
09h00
David DiGregorio, Institut Pasteur
Introductory remarks
09h15
Rémi Monasson, Ecole Normale Supérieure
Integration and multiplexing of positional and contextual information by the hippocampal network
10h00
Jakob Macke, TU Munich and research center ceasar
Bridging the gap between mechanistic and statistical models of neural dynamics
10h45 Coffee break
11h15
Christian Vestergaard, Institut Pasteur
Discovery of computational motifs in experimental neural connectomes using nested randomized reference models
12h00
Elad Schneidmann, Weizmann Institute
Learning the code of large neural populations by random projections
12h45 Lunch break
Keynote Lecture (Chair: David DiGregorio)
14h00
Larry Abbott, Columbia University
Lessons for Machine Learning from a Plastic Recurrent Network in Electric Fish
15h00 Coffee break
Machine learning in neuroscience (Chair: Christophe Zimmer)
15h45
Caswell Barry, University College London
Probing the spatial memory system with deep networks
16h30
Srini Turaga, Janelia Research Campus
From biological neural networks to artificial neural networks, with deep learning
17h15
Marco Zorzi, University of Padova
Deep neural networks for modeling perception and cognition
18h00 Poster session & Apéritif
12th October
Measuring neural networks in real life (Chair: Christoph Schmidt-Hieber)
09h00
Eugenia Chiappe, Champalimaud Research
Visuomotor control of walking during exploration in flies
09h45
Tihana Jovanic, Janelia Research Campus
Mapping neural circuits underlying sensorimotor decisions and sequences in Drosophila
10h30 Coffee break
11h15
Kate Jeffery, University College London
How landmarks inform the sense of direction
12h00 Lunch break
Physics of artificial neural networks (Chair: Jean-Baptiste Masson)
13h30
Carey Priebe, Johns Hopkins University
Structure discovery and exploitation in networks
14h15
Naftali Tishby, Hebrew University
The Information Bottleneck theory of Deep Neural Networks: what do the layers represent?
Location
Address: 25 Rue du Docteur Roux, Paris, France