Name: Ad-hoc talk: Sequential learning on graphs with limited feedback
Start: 2016-12-21T16:30:00+01:00
End: 2016-12-21T17:30:00+01:00
Location: Textkernel B.V. (Amsterdam Office)

We managed to grab one of our friends, Michal Valko, during his visit to Amsterdam to give a talk on some of his recent work. It is quite a bit on a short notice, but that doesn't make you any less welcome to join :)

PLEASE NOTE: Being an ad-hoc talk (on a short notice and during office hours) we won't be offering the usual pizzas and drinks; but you're still very welcome to join if you can! Also, please be respectful of the people working in the office and try to be here at 16:00 sharp. If you arrive too early, we'd suggest to enjoy a relaxing walk in Vliegenbos (otherwise we'll have to seat you in the lobby, and that's no fun).

Talk details

In this talk, we investigate the structural properties of certain sequential decision-making problems with limited feedback (bandits) in order to bring the known algorithmic solutions closer to a practical use including, online influence maximization or sequential recommender systems. To address these structured settings, we can always ignore the graph and use known algorithms for multi-armed bandits. However, their performance scales unfavorably with the number of nodes N, which is undesirable when N means a thousand of sensors or a million of movies. We describe several graph bandit problems and show how to use their graph structure to design new algorithms with faster learning rates, scaling not with N but with graph-dependent quantities, often much smaller than N in real-world graphs.

Speaker bio

http://researchers.lille.inria.fr/~valko/hp/ (https://www.google.com/url?q=http%3A%2F%2Fresearchers.lille.inria.fr%2F~valko%2Fhp%2F&sa=D&ust=1482090052561000&usg=AFQjCNF9Mlr8UERyNQ0wnqKTsl5SXWWxdA)

Mihai Rotaru

Textkernel Talks

Technology

Innovation

Technology Professionals

Tech Talks

New Technology

Web Development

Software Engineering

Big Data

Natural Language Processing

Semantic Technologies

Machine Learning

SaaS (Software as a Service)

Data Mining

Semantic Web

Deep Learning

Ovidiu Curcan

Thomas Hantke

Marc Lambrichs

Lesia Tkacz

Jelle Jan Bankert

Ad-hoc talk: Sequential learning on graphs with limited feedback

Textkernel B.V. (Amsterdam Office)

Share

Textkernel Talks

Ad-hoc talk: Sequential learning on graphs with limited feedback

Textkernel Talks

Details

Members are also interested in