Rating is available when the video has been rented.
This feature is not available right now. Please try again later.
Published on Jan 25, 2017
Intrinsically Motivated Multi-Task Reinforcement Learning with open-source Explauto library and Poppy Humanoid Robot Sébastien Forestier, Yoan Mollard, Damien Caselli, Pierre-Yves Oudeyer, Flowers Team, Inria Bordeaux. 2nd rank at Demonstration Awards, NIPS 2016, Barcelona, Spain, December 6th, 2016.
The corresponding paper is available on arXiv with title "Intrinsically Motivated Goal Exploration Processes with Automatic Curriculum Learning": https://arxiv.org/abs/1708.02190
If you have any question regarding the scientific or technical aspects of the demo, please go to https://redd.it/5q9fnr
=== This demo presents an open-source hardware and software platform which allows non-roboticists researchers to conduct machine learning experiments to benchmark algorithms for autonomous exploration and active learning. Open-source code: https://github.com/sebastien-forestie...
=== It also shows a demo of the Active Model Babbling algorithmic architecture. This is an Intrinsically Motivated Multi-Task Reinforcement Learning algorithm that enables a robot to autonomously set its own goals, and decide which one to target in which order by focusing on those which provide maximal learning progress. The demo also integrates social guidance from humans in real time to drive exploration towards particular objects or actions. In this demo, actions are encoded as dynamic motion primitives with 32 continuous dimensions, and the robot perceives the movement trajectories and state of all objects in an 130 dimensional continuous perceptual space.
=== This project was conducted within a larger long-term research program at the Flowers lab on mechanisms of lifelong learning and development in machines and humans. This research program has in particular lead to a series of novel intrinsically motivated learning algorithms working on high-dimensional real robots and opening new perspectives in cognitive sciences. Papers providing this broader context are:
Oudeyer, P-Y. and Smith. L. (2016) How Evolution may work through Curiosity-driven Developmental Process, Topics in Cognitive Science, 1-11. https://hal.inria.fr/hal-01404334
P-Y. Oudeyer, J. Gottlieb and M. Lopes (2016) Intrinsic motivation, curiosity and learning: theory and applications in educational technologies, Progress in Brain Research, 229, pp. 257-284. https://hal.inria.fr/hal-01404278
Baranes, A., Oudeyer, P-Y. (2013) Active Learning of Inverse Models with Intrinsically Motivated Goal Exploration in Robots, Robotics and Autonomous Systems, 61(1), pp. 49-73. https://hal.inria.fr/hal-00788440
* Poppy Project: an open-source 3D printed low-cost humanoid robotic platform that allows non-roboticists to quickly set up and program robotic experiments. https://www.poppy-project.org
* Explauto: an open-source Python library to benchmark active learning and exploration algorithms that includes already implemented real and simulated robotics setups and exploration algorithms. https://github.com/flowersteam/explauto