The University of Sheffield
Neil Lawrence ML@SITraN
Neil Lawrence

Professor Neil Lawrence

Professor of Machine Learning

Department of Computer Science
Regent Court
211 Portobello
University of Sheffield
Sheffield
S1 4DP

and

Professor of Computational Biology
The Sheffield Institute for Translational Neuroscience
385A Glossop Road
Sheffield
S10 2HQ

Phone: +44 114 xxx xxxx
Fax:     +44 114 222 2290

email: N.Lawrence@dcs.sheffield.ac.uk


Lecture: What is Machine Learning?

Rotation Matrix

News

The next Gaussian Process Summer School will be held in Sheffield from 15th to 17th September 2014. It will be followed on September 18th by a Workshop on Gaussian Processes for Feature Extraction.

Kappenball on the iPhone!

The game of Kappenball is now released on the iPhone! Kappenball is a game illustrating a mathematical theory for procrastination. I introduced it in my inaugural lecture. Kappenball illustrates an effect of stochasticity on decision making. The original example is a stochastic optimal control example described by Bert Kappen at a 2008 workshop I helped organize. The game is designed to illustrate how in a deterministic world we should act on our plans immediately, but in a stochastic world we should often delay action until the right moment. It illustrates why procrastination is sometimes a good thing.

With Mauricio Álvarez we organized a Gaussian Process Road Show in Perreira, Colombia in February.

We ran a Gaussian Process Winter School in Sheffield from Monday 13th January to Wednesday 15th January 2014.

We hosted a 1-day workshop on Gaussian Processes for Spatiotemporal Modelling on Thursday 16th January 2013. It followed directly after the Winter School.

MASAMB 2014 is being hosted by the group in Sheffield. With Nicolo Fusi we will also organise a colocated workshop on statistical genetics and a one day workshop showcasing Sheffield computational biology.

With James Hensman, Joaquin Quinonero Candela and Tianshi Gao, Neil is organising a NIPS workshop on Probabilistic Models for Big Data.

Moving on: September 2013, two researchers from the group are moving on to faculty jobs this month, Nicolas Durrande has returned to France and Ciira Maina has taken a position at Dedan Kimathi University of Technology.

In August we held a Gaussian Process Road Show at the University of Makerere in Kampala, Uganda where we were hosted by John Quinn and Martin Mubangizi.

In June we held a Gaussian Process Summer School in Sheffield.

Neil is Tutorials Chair for NIPS 2013. We've got a great line up of tutorials, available here.

Welcome

In Computer Science Neil leads the Machine Learning Research Group. In Neuroscience he leads the Computational Biology Research Group. His research interests are in probabilistic models with applications in computational biology and personalized health.

CommunityML

Our CommunityML project is about teaching machine learning locally. This includes talks in schools on machine learning related subjects and a mentoring scheme, where local school students spend time in the lab mentored by one of our PhD students. Last year Nicolo Fusi mentored Adam Watts (Notre Dame High School), who used Python to analyse publication records using topic models and this year Max Zwiessele is working with Tim Slater of Airbus to mentor Jake Johnson (King Edward's School) who is looking at the aerodynamics of cycling with Python and Gaussian processes. Neil also taught on the department's Headstart Summer School and will teach in our upcoming Cutler's Ambassador's programme. You can also see a video of Neil's inaugural lecture (targeted at a general audience) here.

MLCommunity

As well as working to bring understanding of machine learning to our local community, the group does a lot of work in the machine learning community to spread ideas and trigger debate, this includes organizing workshops (we will organize 4 in 2014) and summer schools (we will organize two in 2014) as well as teaching on summer schools, reviewing papers, editing journals etc..

Teaching

This year Neil's teaching first year Machines and Intelligence (COM1005 and COM2007) and fourth year/MSc Machine Learning and Adaptive Intelligence (COM6509 and COM4509).

Research Group

MRC Fellow


James Hensman

Research Fellows

Teo de Campos ( RADIANT Project, with Magnus Rattray)
Mu Niu ( BioPreDyn Project, with Magnus Rattray)
Zhenwen Dai (RADIANT Project, with Magnus Rattray)
Javier Gonzalez Hernandez (Recombinant CHO Cell Project, with David James)

PhD Students


Nicoló Fusi
Andreas Damianou
Ricardo Andrade Pacheco
M. Arif Rahman
Alan Saul
Max Zwiessele (MLPM Project)
James McMurray (MLPM Project)

Quotes

discipulus est prioris posterior credulitatis

-Neil Lawrence 2013 (modified, probably incorrectly, from Publilius Syrus c 50 BC)

They say that Understanding ought to work by the rules of right reason. These rules are, or ought to be, contained in Logic; but the actual science of Logic is conversant at present only with things either certain, impossible, or entirely doubtful, none of which (fortunately) we have to reason on. Therefore the true Logic for this world is the Calculus of Probabilities, which takes account of the magnitude of the probability (which is, or which ought to be in a reasonable man's mind). This branch of Math., which is generally thought to favour gambling, dicing, and wagering, and therefore highly immoral, is the only ‘Mathematics for Practical Men’, as we ought to be.

—James Clerk Maxwell in a letter to Lewis Campbell, circa July 1850

pluralitas non est ponenda sine neccesitate

-William of Ockham

The law that entropy always increases, holds, I think, the supreme position among the laws of Nature. If someone points out to you that your pet theory of the universe is in disagreement with Maxwell's equations — then so much the worse for Maxwell's equations. If it is found to be contradicted by observation — well, these experimentalists do bungle things sometimes. But if your theory is found to be against the second law of thermodynamics I can give you no hope; there is nothing for it but to collapse in deepest humiliation.

—Sir Arthur Stanley Eddington, The Nature of the Physical World (1927)

Three Phase Oil Data

The three phase oil data used to be hosted by Aston, but the site seems to be down. I use this data a lot so I've put this site up on the data.

Old News

Inaugural Lecture

Neil's inaugural lecture was on 6th September 2012 at 17:15 in St George's Church Lecture Theatre at the University of Sheffield. The title was ‘Life the Universe and Machine Learning’. More details and an abstract can be found here.

Machine Learning as Engine Design

A blog post on Machine Learning as Engine Design as a response to Kiri Wagstaff's ICML paper.

Teaching 2012-13

This year Neil taught Machine Learning and Adaptive Intelligence (COM6509 and COM4509) module and he was also the staff contact for COM3310 taught by Malcolm Beattie. For information (via Tony Cowling who was staff contact last year) on COM3310 see here.

CVPR 2012 GP Tutorial

With Raquel Urtasun Neil presented a Short Course on Gaussian Processes at CVPR in Providence.

Review of Our Edited Book

A review of the volume on "Learning and Inference in Computational Systems Biology" (edited with Mark Girolami, Magnus Rattrayand Guido Sanguinetti) has been published by Ernst Wit in Biometrics. A further review has been published by Terence Speed in The Quarterly Review of Biology (not open access unfortunately).

Computer Science

A blog post containing personal thoughts on machine learning and Computer Science degrees.

MLSS 2012

I'll be giving three lectures at the Machine Learning Summer School in La Palma. The details of the lectures are available here.

AISTATS 2012

Along with Mark Girolami, Neil is one of the programme chairs for AISTATS 2012. Bernhard Schölkopf is the general chair and Fernando Pérez-Cruz is handling local arrangements.

If you are interested in integrating MATLAB or Octave code into LaTeX, then you might be interested in MATweave. This is a solution for keeping the code you used to create your figures in your LaTeX file.

We've just appointed two post-doctoral research positions associated with our new group in Sheffield.

From 1st August 2010 Neil will be taking up a new position as a collaborative Chair between the departments of Neuroscience and Computer Science at the University of Sheffield. Neil will be based in the Sheffield Institute of Translational Neuroscience. Neil will be joined there by his colleague Magnus Rattray. We will co-lead research groups in machine learning and computational biology.

Book

An edited volume on "Learning and Inference in Computational Systems Biology" is out with MIT Press. Neil edited it with Mark Girolami, Magnus Rattray and Guido Sanguinetti. on Learning in Computational and Systems Biology. It originally emerged from our PASCAL Thematic Programme and the follow on workshops. The MIT Press sites here.

Teaching 2011-12

In Autumn 2011, with Trevor Cohn, Neil taught the Machine Learning and Adaptive Intelligence (COM6509 and COM4509) module and the Introduction to Bioengineering Module (FCE101). See also his Lecture Notes (campus only) page for more details.

AISTATS 2010 in Europe!

For the first time the AISTATS conference was held in Europe. It took place in Sardinia from May 13th-15th 2010. Further European AISTATS are planned for 2012, 2014 etc.. Neil was General Chair for the meeting and Yee Whye Teh and Mike Titterington were the Program Chairs.

Interspeech 2009 Tutorial

For details of an Interspeech tutorial on probabilistic dimensionality reduction see here.

PhD Projects

If you are interested in doing a PhD in Machine Learning you can apply to our research group.

Neil has a number of PhD projects available:

Book

Neil co-edited a volume on dataset shift with Joaquin Quiñonero Candela, Masashi Sugiyama and Anton Schwaighofer. You can find more details on it here.

ICML 2008 Tutorial

For details of an ICML tutorial on probabilistic dimensionality reduction see here.

Thematic Programmes

With Marc Dymetman, Neil organised the first Thematic Programme for the European FP7 Network of Excellence PASCAL II on Leveraging Complex Prior Knowledge for Learning.

With Mark Girolami, Magnus Rattray and Guido Sanguinetti, Neil co-organised a thematic programme for the European FP6 Network of Excellence PASCAL on Learning in Computational and Systems Biology.

Workshops

With Mauricio Álvarez, Neil organized a Statistics and Machine Learning Interface Meeting.

With Joaquin Quiñonero Candela, Neil organised the Bayesian Research Kitchen.

Neil helped out Cédric Archambeau, Ian Roulstone, John Shawe-Taylor and Andrew Stuart in organising Approximate Inference in Stochastic Processes and Dynamical Systems.

Neil helped out Tony Dodd and Rob Harrison in organising a 'Winter School' in Data Modelling.

Neil helped out Matthias Seeger, David Barber and Onno Zoeter in organising a NIPS workshop on "Approximate Bayesian Inference in Continuous/Hybrid Systems".

The Parameter Estimation in Systems Biology PASCAL workshop was co-organised by Neil and Magnus Rattray. It was held in Manchester on the 28-29th March 2007.

Neil helped out Joaquin Quiñonero Candela, Masashi Sugiyama and Anton Schwaighofer in organising a NIPS workshop on "Learning when Test and Training Inputs Have Different Distributions". See the associated book from MIT Press.

Neil co-organised the Gaussian Processes in Practice workshop with Joaquin Quiñonero Candela and Anton Schwaighofer.

Editing

Neil started the JMLR Workshop and Conference Proceedings series and he's currently the series editor, working alongside Mark Reid as the Production Editor.

From October 10th 2011 until he resigned (effective October 10th 2013) Neil was an Associate Editor-in-Chief for IEEE TPAMI.

Neil is an associate editor for JMLR.

Former Post-docs

Nicholas Durrande ( BioPreDyn Project, with Magnus Rattray)
Ciira Maina (Synergy Project, with Magnus Rattray)
Jie Hao (with Magnus Rattray. BioPreDyn Project)
Jaako Peltonen (with Magnus Rattray. Synergy Project)
Michalis Titsias (with Magnus Rattray, TIGRE Project)
Pei Gao (with Magnus Rattray, PUMA Project)
Guido Sanguinetti (worked on PUMA Project)
Huma Lodhi (worked on Human Motion)

Former PhD Students

Alfredo Kalaitzis (viva 3/5/13: now Post-doctoral researcher at UCL with James Nelson, thesis)
Mauricio Álvarez (viva 4/5/11: now Associate Professor at Universidad Tecnológica de Pereira, Colombia, thesis)
Chris Penfold. (viva 27/11/09 with Alastair Goldman): Now at Warwick.
Carl Henrik Ek (viva 25th August 2009 with Phil Torr): Went to Berkeley and now KTH.
Nathaniel King (viva 21/11/2007): Now at IBM.
Yasser Abdel-Haleem (with Steve Renals): Now at IBM.
Tonatiuh Pena Centeno: Now a postdoc at Greifswald.

Computational Health Informatics

Computational Health Informatics.

Systems Medicine

This document last modified Friday, 24-Jan-2014 07:29:11 GMT