Workshop I: From Passive to Active: Generative and Reinforcement Learning with Physics - IPAM

Workshop I: From Passive to Active: Generative and Reinforcement Learning with Physics

Part of the Long Program Machine Learning for Physics and the Physics of Learning
September 23 - 27, 2019

Schedule

All times in this Schedule are Pacific Time (PT)

Monday, September 23, 2019

Morning Session

08:00-08:55
Check-In/Light Breakfast (Hosted by IPAM)
08:55-09:00
Welcome & Opening Remarks: Dean Miguel García-Garibay (Dean of Physical Sciences, UCLA) and Dima Shlyakhtenko (Director, IPAM)
09:00-09:50
Alexandre Tkatchenko (University of Luxembourg)
Towards a Unified Machine Learning Model of Molecular Chemical Space
10:00-10:15
Break
10:15-11:05
Xavier Bresson (National University of Singapore)
Graph Convolutional Neural Networks for Molecule Generation
11:15-11:30
Break
11:30-12:20
Joshua Bloom (University of California, Berkeley (UC Berkeley))
Physics-Informed (and -informative) Generative Modelling in Astronomy

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
Laurent Dinh (Google)
A primer on normalizing flows
15:30-15:45
Break
15:45-16:35
Frank Noe (MLPRC2)
Deep Generative Learning for Physics Many-Body Systems
16:45-17:15
Lightning Poster Presentations
17:15-18:30
Poster Session & Reception (Hosted by IPAM)

Tuesday, September 24, 2019

Morning Session

08:00-09:00
Breakfast (Hosted by IPAM)
09:00-09:50
Philip Kurian (Howard University)
New horizons in quantum biology: Learning complexity, emergence, and coherence in living matter
10:00-10:15
Break
10:15-11:05
Olexandr Isayev (Carnegie Mellon University)
De Novo Drug Design with Deep Reinforcement Learning
11:15-11:30
Break
11:30-12:20
Nicola De Cao (University of Amsterdam)
Deep Generative Models for Molecular Graphs

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
Patrick Riley (Relay Therapeutics)
Reinforcement learning for molecular generation
15:30-16:00
Break
16:00-16:50
Rafael Gomez-Bombarelli (Massachusetts Institute of Technology)
Coarse graining autoencoders and evolutionary learning of atomistic potentials

Wednesday, September 25, 2019

Morning Session

08:00-09:00
Breakfast (Hosted by IPAM)
09:00-09:50
Alán Aspuru-Guzik (University of Toronto)
Generative models for molecules.
10:00-10:15
Break
10:15-11:05
Ross King (University of Science and Technology in Manchester (UMIST))
Automating Science using Robot Scientists
11:15-11:30
Break
11:30-12:20
Katya Scheinberg (Cornell University)
Recent advances in Derivative-Free Optimization and its connection to reinforcement learning.

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
Josh Agar (Lehigh University)
Automatic Feature Extraction from Hyperspectral Imagery using Deep Recurrent Neural Networks
15:30-16:00
Break
16:00-16:50
Lars Ruthotto (Emory University)
A Numerical Analysis Perspective on Deep Neural Networks

Thursday, September 26, 2019

Morning Session

08:00-09:00
Breakfast (Hosted by IPAM)
09:00-09:50
Phiala Shanahan (Massachusetts Institute of Technology)
Generative models for lattice quantum field theory
10:00-10:15
Break
10:15-11:05
Richard Hennig (University of Florida)
Machine-learning for materials and physics discovery through symbolic regression and kernel methods
11:15-11:30
Break
11:30-12:20
Ankit Patel (Rice University)
Breaking Bad: Recent Advances from Function Space Characterization of Neural Nets with Implications for Physical Applications

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
Jascha Sohl Dickstein (Google)
Generalizing Hamiltonian Monte Carlo with Neural Networks
15:30-16:00
Break
16:00-16:15
Joshua Yao-Yu Lin (University of Illinois at Urbana-Champaign)
Hunting for dark matter substructures in strong lensing with neural networks.
16:20-16:35
Nicholas Charron (Rice University)
Coarse-graining molecular systems by spectral matching.
16:40-16:55
Ganesh Sivaraman (Argonne National Laboratory)
A diversified machine learning strategy for predicting and understanding molecular melting points.

Friday, September 27, 2019

Morning Session

08:00-09:00
Breakfast (Hosted by IPAM)
09:00-09:50
Benjamin Nachman (Lawrence Berkeley National Laboratory)
Likelihood free generative modeling for high energy physics
10:00-10:15
Break
10:15-11:05
Tess Smidt (Lawrence Berkeley National Laboratory)
Euclidean Neural Networks* for Emulating Ab Initio Calculations and Generating Atomic Geometries *also called Tensor Field Networks and 3D Steerable CNNs
11:15-11:30
Break
11:30-12:20
Juan Carrasquilla (Vector Institute)
Simulating quantum circuits with neural machine translation

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
Giuseppe Carleo (École Polytechnique Fédérale de Lausanne (EPFL))
Generative and variational modeling for quantum many-body physics
15:30-16:00
Break
16:00-16:50
Han Pu (Rice University)
Solving Quantum Many-Body Problems with Deep Neural Networks
Printable Version