[robotics-worldwide] Call for Papers: ICML Workshop on Planning and Acting with Uncertain Models

Finale P Doshi finale at MIT.EDU
Fri Mar 4 06:27:09 PST 2011


----- First Call for Papers -----

ICML Workshop on Planning and Acting with Uncertain Models
Submission Deadline: April 29, 2011
Workshop Date: July 2, 2011
URL: http://web.mit.edu/~wingated/www/icml2011_workshop

Recent years have seen huge advances in capturing structured uncertainty, 
and many action-selection approaches have been developed to leverage 
knowledge about model-uncertainty: Bayesian approaches cast unknown model 
parameters as additional hidden state; PAC approaches provide guarantees 
on the agent's final performance; robust algorithms minimize overall 
regret. However, simple approximations, such as ignoring much of the 
uncertainty or making myopic decisions, often perform quite well. Does 
their success imply that they are sufficiently sophisticated? In what 
problem domains is explicitly planning to gather information to reduce 
model uncertainty crucial to good performance? What insights do successful 
action-selection strategies provide on how to best track model uncertainty?

The goal of this workshop is to identify state of the art methods for 
planning under uncertainty and discuss why specific methods work well in 
specific domains. For example, how do explicit rewards for reducing 
uncertainty---such as goal-oriented problems---affect what planning 
algorithms are most successful? When planners do not take explicit steps 
to reduce model uncertainty, is that a result of the approximation or the 
fact that, in many domains, most reasonable policies reduce model 
uncertainty reasonably well? When is it useful to introduce internal 
rewards for exploration or reducing model variance? How should agents 
decide what aspects of the model are worth exploring?

Submissions:

We invite full and short papers on recent work exploring these questions. 
Papsers should be in ICML format, submitted to finale at mit.edu by April 29, 
2011.  Notifications for oral and poster presentations will be sent out by 
May 20, 2011.

Program Format:

The format of the workshop will include a series of invited speakers for 
an extended discussion about the various approaches to planning with model 
uncertainty.  Confirmed speakers are Satinder Singh (University of 
Michigan), David Hsu (Stanford University), Joelle Pineau (McGill 
University), and Lihong Li (Yahoo! Research).  The workshop will also 
include a poster session and oral presentations for contributed papers, as 
well as periods to compare different approaches and discuss open problem 
domains.

Organizers:
Finale Doshi-Velez and David Wingate


More information about the robotics-worldwide mailing list