Background

The expanding variety of deep learning methods is proving capable of producing substantive practical gains in a large number of areas of practical machine learning. At the same time, greater understanding of deep learning methods is being developed, through relationship to fundamental formalisms in computer science, relationships to stochastic variational methods, and links to different elements of real neural processing. Deep learning methods hoave provided the capabilities for representation learning that have allowed questions of representation and modelling to be handled at a higher level, with the confidence of not losing too much information in the low level information processing. This workshop will explore the challenges and benefits of using and understanding deep neural networks to ensure continued practical benefits for machine learners, and those who are using machine learning in different domains.

Workshop

On the 9 June 2015 we will be running a second workshop on Deep Learning to bring together those who are working in Machine Learning, Statistics, and the wider community, who are using Deep Machine Learning Methods. We will look at different forms of deep models, and the forms of representations that are learnt, methodological extensions of these methods and the exciting developments of these methods in different areas.

The workshop follows up on the first Edinburgh Deep Learning Workshop in 2014, which had more than 150 attendees.

The workshop will be in the School of Informatics, University of Edinburgh. There will be a number of invited speakers, including Rich Caruana, Alex Graves, Phil Blunsom and Neil Lawrence, but also with room for contributions. The outline of the event is given below, but is subject to change.

• 09.15 Introductory Talk: Amos Storkey.
• 09.30 Invited Speaker: Neil Lawrence (University of Sheffield) on Deep Gaussian Processes.
• 10:15 Pawel Swietojanski (Informatics, Edinburgh) Acoustic Model Adaptation.
• 10.40 Coffee, Discussion
• 11.15 Invited Speaker: Phil Blunsom (University of Oxford and Google Deepmind) on Deep Learning and Computational Linguistics.
• 12:00 Iain Murray (Informatics, Edinburgh) Deep Density Estimation and Scientific Reasoning.
• 12.25 Lunch.
• 13:40 Invited Speaker: Rich Caruana (Microsoft Research) will be examining what we can know of Deep Learning Models.
• 14:25 Michael Pfeiffer (Univ Zurich, ETH Zurich) Deep Spiking Networks.
• 14:50 Coffee, Discussion Period and Posters.
• 15:45 Invited Speaker: Alex Graves (Google Deepmind) on Neural Turing Machines.
• 16:30 Panel Discussion.
• 17:00 End of formal program. Evening pub trip for those who don't need to leave.

• Classifying Plankton Species with Deep Learning and Computer Vision. Matthew Graham, Gavin Gray, Scott Lowe, Finlay Maguire, Alina Selega and Dragos Stanciu
• Learning Grounded Meaning Representations with Autoencoders. Carina Silberer and Mirella Lapata
• Prosodically-enhanced Recurrent Neural Network Language Models. Siva Reddy Gangireddy, Steve Renals, Yoshihiko Nankaku and Akinobu Lee
• Scheduled denoising autoencoders. Krzysztof Geras and Charles Sutton
• A Comparison of Neural Network Methods for Unsupervised Representation Learning on the Zero Resource Speech Challenge. Daniel Renshaw, Herman Kamper, Aren Jansen and Sharon Goldwater
• Bimodal Modelling of Source Code and Natural Language. Miltos Allamanis, Daniel Tarlow, Andrew D. Gordon and Yi Wei
• Chinese Poetry Generation with Recurrent Neural Networks. Xingxing Zhang and Mirella Lapata
• Convolution Neural Networks as controllers for autonomous driving vehicles. Elio Tuci and Aparajit Narayan
• Deep Convolutional Networks for Playing Go. Chris Clark, Amos Storkey, Isaac Henrion

Sponsors

This workshop is funded by the Institute for Adaptive and Neural Computation (ANC), the Scottish Informatics and Computer Science Alliance (SICSA), the Institute for Language, Cognition and Computation (ILCC), and the Centre for Speech Technology Research (CSTR) and ARM.

Costs

The workshop registration is free of charge to registered attendees.

Travel

Participants will need to make their own travel arrangements. By train the nearest station is Edinburgh Waverley, which is less than a 15 minute walk from the forum. See National Rail Enquiries for train information. For those coming from further afield, information about travel to and from Edinburgh Airport is available. A taxi to the city centre from the airport costs about 22GBP to 24GBP one way. There is an express bus from the airport called Airlink that terminates in the city centre and costs 7GBP for a return journey. The journey to the airport requires approximately 30 minutes from the city centre of actual journey time (add a little more during the rush hours). There is also a tram that goes to the centre of town. It takes a little longer than the bus, and is slightly more expensive.

Submission

If you wish to contribute a talk/poster to this workshop, then please send a title and one or two page description (or a paper in conference paper format) to amos+deep@@inf.ed.ac.uk, replacing the dual @ sign. Please send contributions before 31 May (the sooner the better). We will then be in contact about your submission. It is likely that not all submissions will be able to be included.

Registration

If you wish to attend then please register at the eventbrite site. Registration is free.