[Comp-neuro] NIPS 08 Workshop - "Stochastic Models of Behaviour

Aldo Faisal aaf23 at cam.ac.uk
Fri Oct 24 16:56:08 CEST 2008

Call for Abstracts (Poster-session and workshop)

"Stochastic Models of Behaviour" Workshop

at NIPS 2008, Vancouver & Whistler, Canada Important Dates:

    * Submission deadline for poster-session (250 word, plain text):
15th November 2008
    * Rolling notification of acceptance: 3 business days after your submission.
    * Workshop date: 13th December 2008 (Saturday)

URL: http://www.eng.cam.ac.uk/~aaf23/NIPS2008workshop.html
Submission URL and enquiries: smob-nips2008 at googlegroups.com

Advances in experimental methods have increased the availability,
amount and quality of behavioural data for both humans and animals.
Yet most studies lack adequate quantitative methods to model behaviour
and its variability in a natural manner. Thus, a major challenge in
analyzing behavior is to discover some underlying simplicity in a
complex and highly variable stream of actions. The gain of such an
analysis is that the underlying simplicity is often a reflection of
the mechanism driving behavior.
In this workshop we will explore and discuss how to model behaviour
found across many scales of biological organization: From the travel
patterns of people, tools making behaviour, to modeling simple
organisms, such as bacteria, worms, insects - with the ultimate aim to
relate genes and neural circuits to behaviour.
Thus, two major questions emerge:
1. How can we describe and quantify behavior such that its variability
is reflected in a tractable manner?
2. What can we infer from such models of behaviour about the
underlying mechanisms?

We believe that advanced statistical and probabilistic methods (in
which the NIPS conference has been traditionally at the forefront of
developments) can be used to analyze the unconstrained natural
statistics of behaviour. This approach offers a principled route to
extract relevant features of behavioural performance and compare
individual animals in a more objective manner than using highly
constrained experiments. Moreover these models can be generative in
nature providing us with an objective measure to distinguish between
observed behaviours.

The selection of talks is aimed at stimulating ideas and get
discussions started between the participants on how to combine
state-of-the-art machine learning techniques with existing behavioral
approaches to develop a "Bio- informatics of Behavior".

Note: We very much encourage the participation of
experimentally-oriented colleagues working on quantitative data sets
of human, animal or bacterial behaviour and who are interested in
collaboration or inspiration in application of these models.

Organizing committee:
Aldo Faisal, University of Cambridge
Marta C. Gonzalez, Northeastern University

Confirmed, invited speakers
Nathan Eagle (Santa Fe Institute/MIT)
Aldo Faisal (University of Cambridge)
Tony Jebara (Columbia University)
Marta C. Gonzales (Northeastern University)
Greg Stephens (Princeton)

Workshop and poster-session:
 This 1-day workshop session (talks) will serve to get
us going on various approaches towards the two big questions.
There will be plenty of opportunities to interact in small groups prior
and after to the main session.
Participation is open to all. Participants are encouraged to present
their work in the adjoining poster-session (abstract submissions open).

Dr Aldo Faisal
Fellow in Computational
Neuroscience & Biophysics

Dept. of Engineering & Wolfson College
University of Cambridge
Cambridge, UK


More information about the Comp-neuro mailing list