Ellis QPhML 2020

The meeting was held 6-8 July 2020. You can watch the videos of all talks in the links below or on the QPhML2020 YouTube Channel.

The aim of the Ellis program Quantum and Physics based machine learning (QPhML) is to use concepts from Quantum Physics and Statistical Physics to develop novel machine learning algorithms with the ultimate aim to realize novel future, possibly energy efficient, hardware implementations.

The program is part of the recent European initiative called ELLIS (European Laboratory for Learning and Intelligent Systems) to stimulate research on machine learning by building networks of top reseach groups in Europe.

Learn more: [Ellis initiative, Ellis Fellows QPhML Program]

Program

Below you find the tentative program of the workshop. Each talk is given 30 minutes + 10 minutes discussion. After each session, there will be parallel breakout sessions where you can meet with the speakers and discuss further.

Please note that all times are Central European Times (CET)

Day one (06-07-2020)

Opening

Time	Talk	Link
09:45	Riccardo Zecchina and Bert Kappen: “Introduction and welcome”	[video]
09:50	Bernhard Schölkopf (Max Plank Institute for Intelligent Systems, Germany): “Introducing the Ellis Program”	[video]

Machine Learning and Quantum Computing I

Chair: Vedran Dunjko (Leiden University, Leiden)

Time	Talk	Link
10:00	Hans Briegel (Innsbruck University, Innsbruck): “Learning and AI in the quantum domain”	[video, paper]
10:40	Glen Mbeng (Innsbruck University, Innsbruck): “Reinforcement Learning assisted Quantum Optimization”	[video, paper]
11:20	Bert Kappen (Radboud University, Nijmegen): “Learning quantum models from quantum or classical data”	[video, paper]
12:00	Break

Statistical Physics for Machine Learning I

Chair: Lenka Zdeborova (CEA, Saclay)

Time	Talk	Link
13:00	Ton Coolen (Radboud University, Nijmegen): “Replica analysis of overfitting in generalized linear regression models”	[video, paper]
13:40	Manfred Opper, (Technical University, Berlin): “Analysing the dynamics of message passing algorithms using statistical mechanics”	[video, paper]
14:20	Carlo Baldassi (Bocconi University, Milan): “Entropic algorithms and wide flat minima”	[video, paper]
15:00	Break

Machine Learning for Quantum Physics and Complex Systems I

Chair: Riccardo Zecchina (Bocconi University, Milan)

Time	Talk	Link
16:00	Gabor Csanyi (Cambridge University, Cambridge): “Machine learning the quantum mechanics of materials and molecules”	[video, paper]
16:40	Giuseppe Carleo (Flatiron Institute, New York): “Variational Learning of Many-Body Quantum Systems”	[video, paper]
17:20	Juan Carrasquilla (Vector Institute, Toronto): “Neural autoregressive toolbox for many-body physics”	[video, paper]

Day two (07-07-2020)

Connecting Quantum and Machine Learning Theory

Chair: Florian Marquardt (Max Planck Institute for the Science of Light, Erlangen)

Time	Talk	Link
09:20	Jens Eisert (Free University Berlin, Berlin): “Probabilistic Modelling with Tensor Networks - A Bridge from Graphical Models to Quantum Circuits”	[video, paper]
10:00	Masuyuki Ohzeki (Tohoku University, Sendai): “Quantum annealing and machine learning - learning and Black-box optimization”	[video, paper]
10:40	Christian Gogolin (Covestro AG, Leverkusen): “Quantum many-body systems - understanding them with and using them as new machine learning tools”	[video, paper]
11:20	Yunpu Ma and Volker Tresp (Siemens AG and LMU, Munich): “Quantum Machine Learning Algorithms for Knowledge Graphs”	[video, paper]
12:00	Break

Statistical Physics for Machine Learning II

Chair: Gabor Csanyi (Cambridge University, Cambridge)

Time	Talk	Link
13:00	Lenka Zdeborova (CEA, Saclay): “The role of data structure in learning in shallow neural networks”	[video, paper]
13:40	Florian Marquardt (Max Planck Institute for the Science of Light, Erlangen): “Self-learning machines based on nonlinear field evolution”	[video, paper]
14:20	Bert Kappen (Radboud University, Nijmegen): “An atomic Boltzmann machine capable of on-chip learning “	[video, paper]
15:00	Break

Machine Learning and Quantum Computing II

Chair: Bert Kappen (Radboud University, Nijmegen)

Time	Talk	Link
16:00	Roberto Bondesan (Qualcomm AI Research, Amsterdam): “Quantum Deformed Binary Neural Networks”	[video, paper]
16:40	Laurent Daudet (Lighton, Paris): “Mix and Match: leveraging optically-created random embeddings in Machine Learning pipelines”	[video, paper]
17:20	Frank Noe (Freie Universitaet, Berlin): “Deep neural network solution of the electronic Schrödinger equation”	[video, paper]

Day three (08-07-2020)

Machine Learning for Quantum Physics and Complex Systems II

Chair: Manfred Opper, (Technical University, Berlin)

Time	Talk	Link
10:00	Luca Biferale (University of Tor Vergata, Rome): “Equation informed and data-driven tools for data-assimilation and data-classification of turbulent flows”	[video, paper]
10:40	Michele Ceriotti (EPFL, Lausanne): “Symmetry, locality and long-range interactions in atomistic machine learning”	[video, paper]
11:20	Break

Machine Learning for Quantum Chemistry

Chair: Hans Briegel (University Innsbruck, Innsbruck)

Time	Talk	Link
13:00	Alexandre Tkatchenko, (University of Luxembourg, Luxembourg): “Unifying Quantum Chemistry and Machine Learning”	[video, paper]
13:40	Leonard Wossnig (Rahko, London): “Quantum Computing and Machine Learning: How two technologies could enable a new age in chemistry”	[video, paper]
14:20	Anatole von Lilienfeld (University of Basel, Basel): “Quantum Machine Learning in Chemical Compound Space”	[video, paper]
15:00	Break

Machine Learning and Quantum Computing III

Chair: Juan Carrasquilla (Vector Institute, Toronto)

Time	Talk	Link
16:00	Vedran Dunjko (Leiden University, Leiden): “Toward quantum advantages for topological data analysis”	[video, paper]
16:40	Aram Harrow (MIT, Boston): “Small quantum computers and big classical data sets”	[video, paper]
17:20	Andrea Rocchetto (University of Texas, Austin): “The Statistical Limits of Supervised Quantum Learning”	[video, paper]

Organizers

Bert Kappen [web]

Riccardo Zecchina [web]

Gabriele Perugini