Prediction + Optimization 2020 (STOR-i POP)

Theme

Business analytics typically involves tasks of prediction and optimisation. They are usually conducted separately and are not well connected.

The aim of the workshop is to provide platform for exchange of ideas, raise awareness of recent developments and stimulate discussion in the intersection of the areas of prediction and optimisation.

Areas of the workshop

Methodological:

• Data-driven optimization
• Optimisation for machine learning and data science
• Forecasting for optimisation

Applications:

• Inventory and production planning
• Pricing
• Transportation

Workshop dates:

16th and 17th June 2022

Time and Spatial Statistics Workshop 2020 at STOR-i, Lancaster University, UK

The STOR-i Centre for Doctoral Training, which is a joint venture of the Operational Research group at the Department of Management 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 time series and spatial statistics.

The past decades have seen much methodological development in the time series and spatial statistics communities. This is driven both by the increasing volumes and complexity of spatiotemporal data being collected, and the growing number of application domains which benefit from such methods. Despite both time series and spatial statistics being based on ordered sets of observations and hence using analogous analysis techniques, there is often a lack of interaction and dialogue between the two communities.

The main aim of the workshop will be to focus on exploring synergies and overlaps between methodological approaches in time series, statistical signal processing, spatial statistics, and spatiotemporal statistics. There will be a particular focus on topics including:

• modelling nonstationarity and heterogeneity in time and space
• distributional properties of estimators, and techniques for their derivation
• issues surrounding computational scalability and tractability of methodology
• modelling irregular data types (e.g. data that are non-Gaussian, irregularly sampled, contaminated by outliers, observed as point processes)

Important Dates

Friday 17th April 2020: Abstract submission for online talk

Friday 1st May 2020: Registration deadline

Thursday 21^st - Friday 22^nd May 2020: Online Workshop

Confirmed Speakers

Finn Lindgren

University of Edinburgh, UK

Emma McCoy

Imperial College, London, UK

Edith Gabriel

Institut National de la Recherche Agronomique (INRAE), France

Joe Guinness

Cornell University, USA

Qiwei Yao

London School of Economics, UK

Abstract Submission

We invite researchers and practitioners to present their work on time and spatial statistics by submitting an abstract. 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 stor-i-tss@lancaster.ac.uk by Friday 17^th April 2020. All abstracts will be considered for online talks. We apologise if there is not enough capacity in the programme to accept all abstracts. Notifications will be made by Friday 24th April 2020.

Registration

Registration is free. To register, please email stor-i-tss@lancaster.ac.uk by Friday 1st May 2020. Due to possible capacity constraints on the online platform, we ask for a very short 50-word max blurb on why this workshop is of interest to you, which will allow us to restrict participation if necessary. We will confirm registrations by Friday 8th May 2020.

Programme

The workshop will be held on Thursday 21^st and Friday 22^nd May 2020. Delivered online, free registration.

Programme and Organising Committee

Adam Sykulski

Alex Gibberd

Rebecca Killick

Ben Taylor

Statistical Signal Processing (SSP) Workshop 2018 at STOR-i, Lancaster University, UK

The STOR-i Centre for Doctoral Training, which is a joint venture of the Operational Research group at the Department of Management 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 statistical signal processing.

The need for modern statistical signal processing techniques arises due to the ubiquity of vastly-observed signals across numerous applications. Such techniques commonly encompass aspects of time series analysis, machine learning, stochastic processes, optimisation, and many others. Furthermore, signals are commonly observed jointly, motivating the need for methodology that extends to multivariate and high-dimensional settings.

The main aim of the workshop will be to focus on challenges arising in non-parametric methods for non-stationary and high-dimensional signals, bringing together researchers from a number of distinct but related areas of signal processing to share ideas and foster dialogue.

Important Dates

Wednesday 28th February 2018: Abstract submission and student bursary request deadline

Friday 16th March 2018: registration deadline

Thursday 12th - Friday 13th April 2018: Workshop

Invited Speakers

Josiane Zerubia

INRIA, France

Andrew Walden

Imperial College, London, UK

Peter Schreier

Universität Paderborn, Germany

David Simpson

University of Southampton, UK

Peter Craigmile

Ohio State University, USA

Abstract Submission

We invite researchers and practitioners to present their work on statistical signal processing by submitting an abstract. Please send your name, e-mail, affiliation, title and abstract including a list of co-authors in a one-page PDF format by Wednesday 28th February 2018. We will aim to accommodate 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:

• multivariate and multiscale signals
• wavelet and Fourier-based methods
• signal processing on graphs and networks
• applications of statistical signal processing

Student bursaries

We have limited funding to provide bursaries to UK PhD students to enable them to attend the conference. The bursaries are likely to cover accommodation, travel and workshop fees. If you would like to apply please by 28th February 2018 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.

The registration fee is £100 to be paid by Friday 16th March 2018.

Registration for STOR-i students is free, but is required by Friday 9th March 2018.

The conference fee includes access to the workshop presentations, meals and the conference dinner, but you will need to book your own accommodation.

On-campus accommodation can be booked at Lancaster House Hotel or at Lancaster University Visitor Rooms. There are several accommodation options in Lancaster city centre.

Programme

The workshop will be held on Thursday 12th and Friday 13th April 2018.

The programme for the workshop is available here: SSP Workshop Programme

Programme and Organising Committee

Adam Sykulski

Matt Nunes

Rebecca Killick

Ed Cohen

Location

Postgraduate Statistics Centre
Lancaster University
Lancaster
United Kingdom
LA1 4YF

Multi-armed Bandits

The STOR-i Centre for Doctoral Training, which is a joint venture of the Operational Research group at the Department of Management 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.

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

• multi-armed bandit formulations: classic, adversarial, non-parametric, Bayesian, restless, stochastic, infinitely-armed, many-armed, arm-acquiring bandits, best-arm identification, etc.
• theoretical aspects of the exploration-exploitation trade-off 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

We are grateful to the London Mathematical Society for additional support for this workshop, in particular for significantly enhanced support for UK PhD students.

Important Dates

• Friday 20th November 2015: Abstract submission and student fee-waiver request deadline
• Monday 14th December 2015: registration deadline
• Monday 11th - Tuesday 12 January 2016: Workshop

Invited Speakers

• Shipra Agrawal, Columbia University, USA - View Presentation (Agrawal)
• Donald A. Berry, MD Anderson Cancer Center, USA
• Alexandra Carpentier, University of Potsdam, Germany - View Presentation Carpentier
• Kevin Glazebrook, Lancaster University, UK - View Presentation Glazebrook
• Aurélien Garivier, Universite Toulouse, France - View Presentation Garivier
• Katja Hofmann, Microsoft Research, UK
• Michael N. Katehakis, Rutgers University, USA - View Presentation Katehakis
• Emilie Kaufmann, DYOGENE-INRIA, France - View Presentation Kaufmann
• Odalric Maillard, INRIA France
• Yevgeny Seldin, University of Copenhagen, Denmark
• Michal Valko, INRIA-Sequel, France - View Presentation Valko
• Sofia S. Villar, MRC Biostatistics Unit, UK - View Presentation Villar

Contributed Presentations

• Arnoud Den Boer - View Presentation DenBoer
• Hans Degroote - View Presentation Degroote
• James Edwards - View Presentation Edwards
• Andres Gyorgy - View Presentation Gyorgy
• Vladimir Novak - View Presentation Novak
• Faye Williamson - View Presentation Williamson

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 Friday 20th November 2015. We will aim at accommodating all the accepted abstracts as talks, but some may be accepted as posters.

Student bursaries

We have funding to support UK PhD students to attend the workshop. To apply by Friday 20th November 2015 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.

The registration fee is £100 to be paid by Monday 14th December. Registration for STOR-i students is free but is required by Monday 7th December. The conference fee includes access to the workshop presentations, meals and the conference dinner, but you will need to book your own accommodation.

On-campus accommodation can be booked at Lancaster House Hotel or at Lancaster University Visitor Rooms. There are several accommodation options in Lancaster city centre.

Programme

The talks will be held on Monday 11th (10:25 - 17:30) and Tuesday 12th January 2016 (9:30 - 17:35). The poster session will be held on Monday (18:00 - 19:30) followed by a workshop dinner.

Organisers

David Leslie, Steffen Grunewalder, Peter Jacko.

Location

For how to arrive, see the Programme (which also gives more details on the train disruptions between Manchester and Lancaster on the weekend 9-10 January). You can find further information here: http://www.lancaster.ac.uk/contact-and-getting-here/maps-and-travel/

Faculty of Arts & Social Sciences
Lancaster University
Lancaster
United Kingdom
LA1 4YF

Nonstationary Multivariate Time Series

Analysis of nonstationary multivariate time series Workshop 2016 at STOR-i, Lancaster University, UK

The STOR-i Centre for Doctoral Training, which is a joint venture of the Operational Research group at the Department of Management 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 analysis of nonstationary multivariate time series. Lancaster University has a long tradition of research on this topic and several STOR-i PhD students have contributed to it.

Multivariate time series arise in many natural and industrial applications. These series are often high-dimensional in nature and exhibit complex temporal characteristics. Being able to analyse the structure and interplay between such series is key to understanding the dynamics of many important physical processes. Moreover, the availability of high frequency sources has underlined the need for efficient algorithms for analysis of time series in high-dimensional settings.

The main aim of the workshop is to bring together researchers from a number of distinct but related areas of time series analysis to share ideas and foster dialogue to tackle modern challenges in nonstationary multivariate time series analysis.

Important Dates

Friday 26th February 2016: Abstract submission and student bursary request deadline

Friday 18th March 2016: registration deadline

Thursday 14th - Friday 15th April 2016: Workshop

Invited Speakers

David Stoffer

University of Pittsburgh, USA

Hernando Ombao

University of California at Irvine, USA

Rainer Von Sachs

Universit ́e Catholique de Louvain, Belgium

Guy Nason

University of Bristol, UK

Sofia Olhelde

University College London, UK

Abstract Submission

We invite researchers and practitioners to present their work on nonstationary multivariate time series by submitting an abstract. 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 stor-i-nsmvts@lancaster.ac.uk by Friday 26th February 2016. We will aim to accommodate 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:

• multivariate time series modelling
• changepoint analysis
• efficient inference in online data streams
• analysis of networks observed over time
• applications of nonstationary multivariate time series in practice.

Student bursaries

We have limited funding to provide bursaries to UK PhD students to enable them to attend the conference. The bursaries are likely to cover accommodation, travel and workshop fees. If you would like to apply please email stor-i-nsmvts@lancaster.ac.uk by 26th February with a description of your current research topic and why the workshop is important to you.

Registration

To register, please to the registration page.

The registration fee is £100 to be paid by 18th March.
Registration for STOR-i students is free, but is required by Friday 11th March.
The conference fee includes access to the workshop presentations, meals and the conference dinner, but you will need to book your own accommodation.
On-campus accommodation can be booked at Lancaster House Hotel or at Lancaster University Visitor Rooms. There are several accommodation options in Lancaster city centre.

Local Organising Committee

Matthew Nunes, Rebecca Killick, Idris Eckley, Lawrence Bardwell, Jamie-Leigh Chapman, Matt Ludkin.

Location

Postgraduate Statistics Centre
Lancaster University
Lancaster
United Kingdom
LA1 4YF

Multivariate and Spatial Extremes, with Environmental Applications

The STOR-i Centre for Doctoral Training will be hosting a workshop on Multivariate and Spatial Extremes at Lancaster University from 4-6th July 2016. As well as welcoming presentations on theoretical and methodological aspects, we are also hoping to have a number of talks on uses of extreme value theory in complex environmental applications.
Research into extreme value theory is well-established at Lancaster. Currently, the extremes group consists of three members of staff (Professor Jonathan Tawn, Dr Emma Eastoe and Dr Jenny Wadsworth), five STOR-i PhD students and one RA. We have strong links with a number of partners, both academic and industry.

Important Dates

• End of April 2016: Student fee-waiver request deadline
• Mid-May 2016: Abstract submission deadline
• Beginning of June 2016: Deadline for registration and payment of registration fee
• July 4th-6th 2016: Workshop

Invited Speakers

• Simon Brown, UK Met Office
• Anthony Davison, Ecole Polytechnique Federale de Lausanne
• Theo Economou, University of Exeter
• Phil Jonathan, Shell
• Ben Marchant, British Geological Survey
• Thomas Opitz, INRA
• Anne Sabourin, Telecom Paris-Tech
• Johan Segers, Universite Catholique de Louvain
• Ben Shaby, Penn State University
• Jonathan Tawn, Lancaster University

Abstract Submission

Abstract submission for contributed talks has closed, but we continue to invite submissions for posters.

To submit your abstract, 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 stor-i-extremes@lancaster.ac.ukby the registration deadline of Friday 3 June.

Within the themes of multivariate and spatial extremes, some topics that we hope to cover include:

• Max-stable processes;
• Conditional extremes;
• Hierarchical models;
• Geostatistical-type models;
• Environmental applications of extremes: air pollution, climate models, river flows, wave heights, precipitation,...

We are particularly interested in novel applications, even if you haven't yet developed a fully-fledged modelling strategy.

Student Fee Waiver

We have limited funding to waive the workshop fees of UK PhD students. If you would like to apply for a fee waiver please email stor-i-extremes@lancaster.ac.uk by the end of April 2016 with the name of your supervisor(s), a description of your research topic and why the workshop would be beneficial to you.

Registration

The registration fee is £100 to be paid by Friday 3 June 2016.

The registration fee includes access to the workshop presentations, coffee breaks, lunches on Tuesday and Wednesday, and the workshop dinner on Tuesday evening.

Travel and accommodation details

Although a more detailed timetable will be available closer to the workshop, in order to help plan your travel arrangements please note that registration will be from at 13.00 on Monday 4th July and the workshop will close after lunch on Wednesday 6th July. Accommodation on-campus can be found at the Lancaster House Hotel, or through the university room booking service. There are a number of options in Lancaster city centre, including Lancaster Central Travelodge and the Kings Arms Hotel. The town centre is approximately 3 miles (5 km) from the university campus. Whilst it is possible to walk between the two, buses from the town centre to campus run frequently.

Organisers

Emma Eastoe, Jenny Wadsworth

Location

Postgraduate Statistics Centre
Lancaster University
Lancaster
United Kingdom
LA1 4YF

Decision Theory and Network Science: Methods and Applications

There is a wide array of problems in applied science, business analytics and industry that involve a decision-making process that takes into account network structure. The goal of the event is to encourage the exchange of ideas and seed collaboration among researchers developing methodology in the intersection of decision theory and networks Science, bringing together statistics and management science.

Date:18th of September, 2017

Location: A54 LT, PSC

Confirmed speakers:

• Michal Valko (Inria, France)
• Sergio Bacallado (University of Cambridge)
• Wolfram Wiesemann (Imperial College)
• Matthias Müller-Hannemann (University of Halle-Wittenberg)
• Matthew Nunes (Lancaster University).

Three more speakers will be added to the programme.

Local organisers:

• Simón Lunagómez (Department of Mathematics and Statistics)
• Marc Goerigk (Management Science)