Alan Turing Institute Deep Learning Scoping Workshop

Edinburgh, 23-24 November 2015.

See the ATI Deep Learning Open Workshop for details for 25 November.

Ground floor, Informatics Forum, 10 Crichton Street, Edinburgh
Directions to the Informatics Forum are available

Local Organisers: Amos Storkey, Krzysztof J. Geras


Summary description

Deep Learning is the learning and use of highly flexible multilayer models for machine learning. Deep learning has proven itself, both in terms of academic research and capability, and in terms of commercial interest. The Deep Learning Scoping Workshop will work on establishing areas of joint research and furthering UK capacity and capability building in this field. It will link established and junior researchers, build links with commercial research groups and introduce members to new problems and data sources. The aim of the workshop is to ensure UK researchers are on the crest of the wave of deep learning. We will plan for an ATI theme in deep learning, agree a process on early and regular communication, establish common research themes, and propose potential funding bids.


Key scientific questions to be answered

The workshop will focus on four aspects of deep learning. First, we will consider the further generalisations of deep learning methods beyond classification and feature generation, to areas of generative modelling, transfer learning, multimodal integration, reinforcement learning and other important problems. Second, we will ask what other practical problems/scenarios can benefit from the application of deep learning methods beyond current domains, and how models can be distilled and made accessible for use in those domains. Third, we will propose potential approaches for reducing or distributing the computational costs associated with deep learning. Fourth we will examine the theoretical aspects of deep learning, and ask why deep learning has proven so effective and what the limitations are.


Key additional topics to be addressed

In addition to addressing the key scientific questions above, we will discuss the potential benefits of coordinated activity. The topics of this discussion will include
  • Facilitating trusted early communication of approaches and results to enable the UK to maintain its position in the fast and competitive research in this field.
  • The potential to maintain joint data repositories that will provide benefits to all parties.
  • The steep learning curve for doctoral students and postdocs. The working environment is quickly changing. What coordination will helps ensure researchers are quickly up to speed?
  • What interaction is required with business research labs and startups to enable them to take up deep learning methods, where otherwise they would not have the opportunity.
  • What the best practice is, in training the next generation in deep learning.

Workshop

The ATI Workshop on the 23th and 24th is for invitees involved with scoping the focus of deep learning research in the Alan Turing Institute. The workshop on the 25th is open to everyone.

The workshop will be in the School of Informatics, University of Edinburgh.

The workshop plan is given below but is subject to change

Mon 23rd Nov

  • 9.30 Start and Introduction. Amos Storkey, University of Edinburgh
  • 9.40 Speaker: Yee Whye Teh, University of Oxford.
    Interface between Monte Carlo/sampling methods and optimization via variational inference.
  • 10.15 Speaker: Zhenwen Dai, University of Sheffield.
    Variational Auto-encoded Deep Gaussian Processes
  • 10.35 3 Minute Talks. Pawel Swierojanski, Iain Murray, James Wilson, Ben Graham
  • 11:00 Coffee
  • 11.30 Speaker: Andrew Zisserman, University of Oxford + Google DeepMind.
    Comments on spatial transformers, and on non-classification applications of CNNs.
  • 12.00 3 Minute Talks. Steve Renals, Nicholas Heess, Mirella Lapata, Philip Blunsom, Tom Nickson
  • 12:30 Lunch
  • 1:30 Speaker: Zoubin Ghahramani, University of Cambridge
    Deep learning and probabilistic models.
  • 2:00 Breakout: Beyond classifiers. Best applications. The value of Data Repositories.
  • 3:20 Coffee
  • 4:00 Speaker: Yarin Gal, University Cambridge,
    Modern deep learning through Bayesian eyes
  • 4:30 3 Minute Talks. Emre Ozer, Lukasz Romaszko, Harri Edwards, Ali Eslami, Kai Arulkumaran
  • 5:00 Breakout: Computational costs in deep learning. Interactions between Universities companies.
  • 6:00 Drinks and Discussion.
  • 7.30 Blonde Restaurant.

Tues 24th Nov

  • 9.15 Reconvene for day 2
  • 9.30 Speaker: Andrea Vedaldi, University Oxford
    Understanding visual representations.
  • 10.00 Speaker: Krzysztof Geras, University of Edinburgh
    Compressing LSTMs into CNNs
  • 10.30 3 Minute Talks. Andreas Damianou, Taco Cohen, Amos Storkey, Karl Moritz Herman, Andrew Stuart.
  • 11.00 Coffee
  • 11.30 Speaker: Yann Ollivier, CNRS
    Training recurrent networks without backtracking in time.
  • 12:00 Breakout: Training and Coordination.
  • 12:30 Lunch
  • 1:30 Speaker: Michael Pfeiffer, University of Zurich, ETH Zurich.
    Deep spiking neural networks – low-latency, low-compute classifiers for neuromorphic platforms.
  • 2:00 3 Minute Talks. David Reichert, Gavin Gray, Alison Lowndes
  • 2:20 Breakout: What theory does Deep Learning need?
  • 3:20 Coffee
  • 4:00 Speaker: Nando De Freitas, University of Oxford + Google DeepMind.
    Deep Learning: scaling, programming and acting
  • 4:30 Speaker: Guillame Bouchard, University College London.
    Machine Reading by Learning to Update
  • 4:50 Discussion and ATI reporting.
  • 7pm Dinner at Angels with Bagpipes.

Wed 25th

Please see the ATI Deep Learning Open Workshop pages for Wednesday's programme.

Travel

Participants will need to make their own travel arrangements. Hotels will be booked by the workshop. 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.

Please also join our Facebook group