CALL FOR PAPER

IEEE Transactions on Autonomous Mental Development,
special issue on Active learning and intrinsically motivated exploration in robots

http://www.ieee-cis.org/pubs/tamd/
http://flowers.inria.fr/tamd-activeLearningIntrinsicMotivation.htm

This special issue is jointly supported by the
IEEE CIS Technical committee on Autonomous Mental Development, http://research.microsoft.com/en-us/um/people/zhang/amdtc/
and the IEEE RAS Technical committee on Robot Learning, http://www.learning-robots.de/

Learning techniques are increasingly being used in todays’ complex robotic system. Robots are expected to deal with a large variety of tasks, using their high-dimensional and complex bodies, to interact with objects and humans in an intuitive and friendly way. In this new setting, not all relevant information is available at design time, thus self-experimentation and learning by interacting with the physical and social world is very important to acquire knowledge.

A major obstacle, in high and complex sensorimotor space, is that learning can become extremely slow or even impossible without adequate exploration strategies. To solve this problem, two main approaches are now converging. Active learning, from statistical learning theory, where the learner actively chooses experiments in order to collect highly informative examples, and where expected information gain can be evaluated with either theoretically optimal criteria or various computationally efficient heuristics. The second approach, intrinsically motivated exploration, from developmental psychology and recently operationalized in the developmental robotics community, aims at building robots capable of open-ended cumulative learning through task-independent efficient exploration of their sensorimotor space and to refine our understanding of how children learn and develop.

Although similar in some aspects, these two approaches differ in some of the underlying assumptions. Active learning implicitly assumes that samples with high uncertainty are the most informative and focuses on single tasks. On the contrary, Intrinsic motivation has been identified by psychologists as an innate incentive that pushes organisms to spontaneously explore activities or situations for the sole reason that they have a certain degree of novelty, challenge or surprise, hence the term curiosity-driven learning sometimes used.

Several open problems exist still and the goal of this special issue is to show state-of-the-art approaches to these problems and open new directions. Papers should address the following, non-exhaustive, topics applied to robotics or animal cognitive model:

• How can traditional active learning heuristics be applied to robotics problems such as motor learning, affordance learning or interaction learning?
• How to select an active strategy ? Are there general purpose methods or are they task dependent?
• How can active and intrinsic motivated exploration enable long-life, task-independent learning and development?
• Is there a unified formalism to both approaches?
• How precisely do they model human active learning and exploration and its role in development?
• Can these approaches be used for social tasks, e.g. joint-work and human-robot interaction ?

Editors:

Manuel Lopes, University of Plymouth, http://www.plymouth.ac.uk/staff/mlopes
Pierre-Yves Oudeyer, INRIA, http://www.pyoudeyer.com

Two kinds of submissions are possible:

• Regular papers, up to 15 double column pages ;
• Correspondence papers either presenting a "perspective" that includes insights into issues of wider scope than a regular paper but without being highly computational in style or presenting concise description of recent technical results, up to 8 double column pages ;

Instructions for authors :

http://ieee-cis.org/pubs/tamd/authors/

We are accepting submissions through Manuscript Central at :

http://mc.manuscriptcentral.com/tamd-ieee (please select « Active Learning and Intrinsic Motivation » as the submission type)

When submitting your manuscript, please also cc it to manuelcabidolopes@gmail.com and pierre-yves.oudeyer@inria.fr

Timeline :

15 Feb 2010 – Deadline for paper submission
30 March – Notification
30 April – Final version
15 May – Electronic publication
30 June – Printed publication