Past Industrial and Interdisciplinary Workshops

9 June 2017
10:00
Stephane Chretien
Abstract

The National Physical Laboratory is the national measurement institute. Researchers in the Data Science Division analyse various types of data using mathematical, statistical and machine learning based methods. The goal of the workshop is to describe a set of exciting mathematical problems that are of interest to NPL and more generally to the Data Science community. In particular, I will describe the problem of clustering using minimum spanning trees (MST-Clustering), Non-Negative Matrix Factorisation (NMF), adaptive Compressed Sensing (CS) for tomography, and sparse polynomial chaos expansion (PCE) for parametrised PDE’s.

  • Industrial and Interdisciplinary Workshops
19 May 2017
10:00
Abstract

Conventional neutron reflection is a very powerful tool to characterise surfactants, polymers and other materials at the solid/liquid and air/liquid interfaces. Usually the analysis considers molecular layers with coherent addition of reflected waves that give the resultant reflected intensity. In this short workshop talk I will illustrate recent developments in this approach to address a wide variety of challenges of academic and commercial interest. Specifically I will introduce the challenges of using substrates that are thick on the coherence lengthscale of the radiation and the issues that brings in the structural analysis. I also invite the audience to consider if there may be some mathematical analysis that might lead us to exploit this incoherence to optimise our analysis. In particular, facilitating the removal of the 'background substrate contribution' to help us focus on the adsorbed layers of most interest.

  • Industrial and Interdisciplinary Workshops
5 May 2017
10:00
Apala Majumdar
Abstract

Liquid crystals are classical examples of mesophases or materials that are intermediate in character between conventional solids and liquids. There are different classes of liquid crystals and we focus on the simplest and most widely used nematic liquid crystals. Nematic liquid crystals are simply put, anisotropic liquids with distinguished directions and are the working material of choice for the multi-billion dollar liquid crystal display industry. In this workshop, we briefly review the mathematical theories for nematic liquid crystals, the modelling framework and some recent work on modelling experiments on confined liquid crystalline systems conducted by the Aarts Group (Chemistry Oxford) and experiments on nematic microfluidics by Anupam Sengupta (ETH Zurich). This is joint work with Alexander Lewis, Peter Howell, Dirk Aarts, Ian Griffiths, Maria Crespo Moya and Angel Ramos.
We conclude with a brief overview of new experiments on smectic liquid crystals in the Aarts laboratory and questions related to the recycling of liquid crystal displays originating from informal discussions with Votechnik ( a company dealing with automated recycling technologies , http://votechnik.com/).
 

  • Industrial and Interdisciplinary Workshops
3 March 2017
10:00
Steve Hilditch
Abstract

Road travel is taking longer each year in the UK. This has been true for the last four years. Travel times have increased by 4% in the last two years. Applying the principle finding of the Eddington Report 2006, this change over the last two years will cost the UK economy an additional £2bn per year going forward even without further deterioration. Additional travel times are matched by a greater unreliability of travel times.

Knowing demand and road capacity, can we predict travel times?

We will look briefly at previous partial solutions and the abundance of motorway data in the UK. Can we make a breakthrough to achieve real-time predictions?

  • Industrial and Interdisciplinary Workshops
9 December 2016
10:00
Alex Codd
Abstract

As part of a suite of products that provide a drive thorough vehicle tyre inspection system the assessment of wheel alignment would be useful to drivers in maintaining their vehicles and reducing tyre wear.  The current method of assessing wheel alignment involves fitting equipment to the tyre and assessment within a garage environment. 

The challenge is to develop a technique that can be used in the roadway with no equipment fitted to the vehicle.  The WheelRight equipment is already capturing images of tyres from both  front and side views.  Pressure sensors in the roadway also allow a tyre pressure footprint to be created.  Using the existing data to interpret the alignment of the wheels on each axle is a preferred way forward.

  • Industrial and Interdisciplinary Workshops
2 December 2016
10:00
Christian Sommeregger & Wen Wong
Abstract

Hotels.com is one of the world’s leading accommodation booking websites featuring an inventory of around 300.000 hotels and 100s of millions of users. A crucial part of our business is to act as an agent between these two sides of the market, thus reducing search costs and information asymmetries to enable our visitors to find the right hotel in the most efficient way.

From this point of view selling hotels is one large recommendation challenge: given a set of items and a set of observed choices/ratings, identify a user’s preference profile.  Over the last years this particular problem has been intensively studied by a strongly interdisciplinary field based on ideas from choice theory, linear algebra, statistics, computer science and machine learning. This pluralism is reflected in the broad array of techniques that are used in today’s industry applications, i.e. collaborative filtering, matrix factorization, graph-based algorithms, decision trees or generalized linear models.

The aim of this workshop is twofold.

Firstly we want to give some insight into the statistical modelling techniques and assumptions employed at hotels.com, the practical challenges one has to face when designing a flexible and scalable recommender system and potential gaps between current research and real-world applications.

Secondly we are going to consider some more advanced questions around learning to rank from partial/incomplete feedback (1), dealing with selection-bias correction (2) and how econometrics and behavioral theory (eg Luce, Kahneman /Tversky) can be used to complement existing techniques (3).

 

  • Industrial and Interdisciplinary Workshops

Pages