Computational Learning and Memory Group

home | members | publications | events | teaching | vacancies | directions

Advances in Statistical Inference in the Visual Cortex

The workshop will consist of short presentations, with the main emphasis being on round-table discussions about current issues in understanding visual cortex as implementing statistical inference. Sample issues: necessary ingredients of a generative model for vision, hierarchical extensions, acceptable approximations of a recognition model, correspondence between neural activity and statistical inference, parametric (PPC) vs sampling-based neural representations, learning and learnability, role of top-down / lateral / recurrent connections, incorporating attention, design of psychophysical / electrophysiological experiments to test theories.

Participants

Pietro Berkes (Brandeis)
Sophie Deneve (ENS)
Jozsef Fiser (Brandeis)
Peter Latham (Gatsby) [tbc]
Mate Lengyel (Cambridge)
Jorg Lucke (FIAS)
Gergo Orban (Cambridge)
Constantin Rothkopf (FIAS)
Cristina Savin (FIAS)
Peggy Series (Edinburgh)
Rich Turner (Cambridge)
Balazs Ujfalussy (RMKI)

Dates

12-13 July 2010

Venue

Central European University
9 Nador utca, Budapest 1051

Accommodation

City Hotel Pilvax, ~67 EUR / night
Hotel Central Basilica, ~72 EUR / night

Travel info

For a list of airlines flying to Budapest and other airport-related information, see the web page of Budapest Airport.

Info on Hungary here.

Info on Budapest here.

Finances

There is no registration fee. Participants cover all their expenses (flight tickets, accommodation, meals, etc).

Programme

12 June, Monday

09.00-12.00: 10 mins presentation + 10 mins discussion / participant: Identifying topics
12.00-14.00: lunch break
14.00-18.00: discussions
18.00-: dinner + submergence in Budapest night life...

13 June, Tuesday

08.30-13.30: discussions
13.30-15.00: lunch break
15.00-17.30: discussions
18.00- dinner + closing banquet + farewell party + faring away well

As you can see, according to the workshop abstract, most of the time will be devoted to discussions. This means that the short presentations of the first morning session will set the themes around which these discussion will be organized. Our aim is to have intensive scientific debate to reveal controversies in the field and, hopefully, reconcile at least some of them. If you make a presentation, it should not be about your results. It should be about the questions you consider most relevant to the title and perhaps a pointer as to how your work could be related to those topics.