Connectionists: ICML Workshop on Markets, Mechanisms, and Multi-Agent Models

[Amos Storkey]

Dear Connectionists

Please see below for a call for contributions for an ICML workshop on 
Markets, Mechanisms, and Multi-Agent Models

The remit for this call is broad, and one particular interest is work 
looking at connections between organisational mechanisms for neural 
systems and organisational mechanisms (such as markets) for economic 
systems. At face value, both have much in common - both involve or 
facilitate some levels of self-organisation, specialisation of function, 
adaptability, distributed computation etc. Recent work has developed on 
the parallels between standard approaches for learning/inference and 
market mechanisms in information markets. We would be very interested in 
contributions which make substantive comment relating economic 
organisation and neural processing and organisational principles.

regards

Amos Storkey



********************************************************************

                     CALL FOR CONTRIBUTIONS

     ICML Workshop on Markets, Mechanisms, and Multi-Agent Models:
     Examining the Interaction of Machine Learning and Economics

       Edinburgh: June 30 or July 1, 2012 (to be determined)

             http://icml2012marketswkshop.pbworks.com/

********************************************************************
Important Dates:
----------------
Deadline for submissions: 20 April 2012
Notification of Acceptance: 18 May 2012

Organisers:
----------
Amos Storkey (a.storkey at ed.ac.uk)
Jacob Abernethy (jaber at seas.upenn.edu)
Jenn Wortman Vaughan (jenn at cs.ucla.edu)

Overview:
---------
Many of society’s greatest accomplishments are in large part due to the 
facility of markets. Markets and other allocation mechanisms have become 
necessary tools of the modern age, and they have been key to 
facilitating the development of complex structures, advanced 
engineering, and a range of other improvements to our collective 
capabilities. Much work in economics has been done to demonstrate that 
markets can, in aggregate, function very well even when the individual 
participants are noisy, irrational or myopic.

In terms of aims and benefits, the design of machine learning techniques 
has much in common with the development of market mechanisms: 
information aggregation, maximal efficiency, scalability, and, more 
recently, decentralization. Current machine learning algorithms are 
often single goal methods, built from simple homogeneous units by one 
person or individual groups. Perhaps looking to the organisations of 
economies may help in moving beyond the current centralised design of 
most machine learning methods. Allowing agents with different opinions, 
approaches or methods to enter and leave the market, to interact, and to 
adapt to changes can have many benefits. For example it may enable us to 
develop methods that provide continuous improvement on complex problems, 
reuse results by improving on previous outcomes rather than building 
bigger models from scratch, and adapting to changes.

There are many relationships between machine learning methods, Bayesian 
decision theory, risk minimisation, economics, statistical physics and 
information theory that have been known for some time. There are also 
many open questions regarding the full nature and impact of these 
connections. This workshop will explore these connections from many 
different directions.

Various Topics:
---------------
More detailed descriptions of each of these topics can be found on the 
website.

1) Prediction markets as a tool for learning and aggregation.
2) Learning in problems of mechanism design.
3) Prediction and learning in ad auctions.
4) Online trading, portfolio selection, etc. in financial engineering.
5) Relating Market Mechanisms and Machine Learning/Neural Methods.
6) Transactional Communication in Multi-agent Systems.

Feel free to email the organizers regarding additional topics.

Submission Instructions:
-----------------------
We are soliciting contributions for talks and for posters. Submissions 
should take the form of a abstract limited to 4 pages plus references. 
At least one page of this should be dedicated to describing the 
relationship of this work to other work in both Economics/Finance and in 
this area of Machine Learning.

In addition if you wish to be considered for a talk, you should submit a 
further description of what the motivation and content of your talk will 
be (in one page or less).

Please see the website for full submission instructions.

-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.