The Power of Graphs in Machine Learning and Sequential Decision Strategies
University College London, UCL, London, UK (https://graphpower.inria.fr)
Workshop Date: June 3-4, 2019
Registration: free
This workshop will bring together researchers aimed at answering this need from a common perspective: applying the tenets of graph signal processing to online sequential decision strategies (and machine learning at large).
Invited Speakers
Keynote: Massimiliano Pontil, UCL
Keynote: Peter Battaglia, Deepmind, Deep learning on graphs
* Aurélien Bellet, Inria, Communication-efficient and decentralized multi-task boosting while learning the collaboration graph
* Richards Combes, Centrale-Supelec, Computationally efficient estimation of the spectral gap of a Markov chain
* Mark Herbster, UCL, Predicting switching graph labelings with cluster specialists
* Pierre Borgnat, CNRS, Optimal transport under regularity prior for domain adaptation between attributed graphs
* Ricardo Da Silva, UCL, Flexible probabilistic models of networks in sequence data and in semi-supervised classification
* Marc Lelarge, Inria, Embeddings for graph classification
* Michael Bronstein, Imperial College London
* Olga Klopp, ESSEC, Sparse network estimation
* Sofia Charlotta Olhede, UCL
* Mirco Musolesi, UCL, Towards decentralized reinforcement learning architectures for modeling agent societies
* Csaba Szepesvári, Deepmind, New insights to partial monitoring
* Quentin Berthet, University of Cambridge
* Varun Kanade, University of Oxford
Call for Posters
We accept submission of 1-page extended abstracts that will be lightly reviewed based on relevance. Accepted contributions will be arranged in the format of poster presentations during the lunch slot on day 2. The workshop is a venue to share recent research results with no published proceedings.
We accept submission related to the future path of the vibrant field of graph signal processing to online sequential decision strategies (and machine learning at large). Example of topics of interests are:
* sequential decision-making on and with graphs
* graph representation for online learning
* graph networks
* geometric deep learning
* active learning graphs with limited information
Email the document to l.toni@ucl.ac.uk as an attachment. The email must have the subject: “UCL Graph Learning Poster” and wait for the decision.
Best regards,
Laura Toni and Michal Valko