Multi-armed Bandit Workshop 2016 at STOR-i, Lancaster University, UK

Dates: 11-12 January 2016

Webpage: http://www.stor-i.lancs.ac.uk/research/Workshops/Multi-armed-Bandit-Workshop

The STOR-i Centre for Doctoral Training, which is a joint venture of the Operational Research group at the Department of Managment Science and the Statistics group at the Department of Mathematics at Lancaster University with strong links to the industry, will be holding a specialist workshop on multi-armed bandits. Lancaster university has a long tradition of research on this topic and several STOR-i PhD students have contributed to it.

The multi-armed bandit problem has become classic because of its extreme difficulty. Researchers from different fields, especially statistics, operational research, applied probability and computer science, have made theoretical contributions that in the recent years started gaining wider popularity in practice due to the huge modelling power of this problem.

This workshop will provide a forum for stimulating discussions, especially across the approaches that have developed independently of each other. The list of invited speakers includes recognised experts on reinforcement learning bandits, restless bandits, non-parametric bandits, Bayesian bandits, etc., covering both theoretical advances and applications in information technology, business & management, communications networks, health care, etc.

We are grateful to EPSRC for support of the series of specialed workshops at STOR-i and to the London Mathematical Society for additional support for this workshop, in particular for significantly enhanced support for UK PhD students.

Important Dates

Monday 23 November 2015: Abstract submission and student bursaries request deadline

Monday 14 December 2015: Registration deadline

Monday 11 - Tuesday 12 January 2016: Workshop

Invited Speakers (confirmed)

Shira Agarwal - Columbia University, USA
Alexandra Carpentier - University of Potsdam, Germany
Chris Gamble - Google UK
Aurelian Garivier - University de Toulouse, France
Kevin Glazebrook - Lancaster University, UK
Katja Hofmann - Microsoft UK
Michael N. Katehakis - Rutgers University, USA
Emilie Kaufmann - DYOGENE-INRIA, France
Odalric Maillard - INRIA France
Remi Munos - Google DeepMind, UK
Yevgeny Seldin - University of Copenhagen, Denmark
Sofia S. Villar - MRC Biostatistics Unit, UK

Abstract Submission

We invite researchers and practitioners to present their work on multi-armed bandits by submitting an abstract below. Please, send your name, e-mail, affiliation, title and abstract including a list of co-authors in a one-page PDF format by e-mail to [log in to unmask] by Monday 23 November 2015. We will aim at accommodating all the accepted abstracts as talks, but some may be accepted as posters.

The topics of the workshop include, but are not limited to, the following:

- multi-armed bandit formulations: classic, adversial, non-parametric, Bayesian, restless, stochastic, infinitely-armed, many-armed, arm-aquiring bandits, best-arm identification, etc.
- theoretical aspects of exploration–exploitation tradeoff in sequential experimental design
- optimality and convergence of randomised algorithms, index rules, and myopic policies
- characterisation, computation, approximation and learning of (near-)optimal policies
- applications of multi-armed bandit models in practice

Student bursaries

We have funding to support UK PhD students to attend the workshop. To apply, please email [log in to unmask] by Monday 23 November with a description of your current research topic and why the workshop is important to you.

Registration

To register, please go to the registration page: http://online-payments.lancaster-university.co.uk/browse/extra_info.asp?compid=1&modid=1&deptid=6&catid=464&prodid=3043 .

The registration fee is £100 to be paid by Monday 14th December.

Programme

The workshop will be held on Monday 11th and Tuesday 12th January 2016, from 9:00 - 18:00. The workshop dinner will be held on Monday.

Detailed program of accepted presentations will be announced in early December 2015

Organisers

David Leslie, Steffen Grunewalder, Peter Jacko.

Location

Postgraduate Statistics Centre
Lancaster University
Lancaster
United Kingdom
LA1 4YF


--
David Leslie, Professor of Statistics,
Department of Mathematics and Statistics, Fylde College
Lancaster University, Lancaster, LA1 4YF, United Kingdom
[log in to unmask]
http://www.lancaster.ac.UK/fas/maths/people/david-leslie


You may leave the list at any time by sending the command

SIGNOFF allstat

to [log in to unmask], leaving the subject line blank.