Guest Talk: Sao Mai Nguyen

20 February 2015
Begin time: 
CITEC 1.015

A Curious Robot Learner for Interactive Goal-Babbling

The challenges posed by robots operating in human environments on a daily basis and in the long-term point out the importance of adaptivity to changes which can be unforeseen at design time. The robot must learn continuously in an open-ended, non-stationary and high dimensional space. It must be able to know which parts to sample and what kind of skills are interesting to learn. One way is to decide what to explore by oneself. Another way is to refer to a mentor. We name these two ways of collecting data sampling modes. The first sampling mode correspond to algorithms developed in the literature in order to autonomously drive the robot in interesting parts of the environment or useful kinds of skills. Such algorithms are called artificial curiosity or intrinsic motivation algorithms. The second sampling mode corresponds to social guidance or imitation where the teacher indicates where to explore as well as where not to explore. Starting from the study of the relationships between these two concurrent methods, we ended up building an algorithmic architecture with a hierarchical learning structure, called Socially Guided Intrinsic Motivation (SGIM).

We have built an intrinsically motivated active learner which learns how its actions can produce varied consequences or outcomes. It actively learns online by sampling data which it chooses by using several sampling modes. On the meta-level, it actively learns which data collection strategy is most efficient for improving its competence and generalising from its experience to a wide variety of outcomes. The interactive learner thus learns multiple tasks in a structured manner, discovering by itself developmental sequences.