Task-Agnostic Morphology Evolution (TAME) evolved morphologies that match multi-task performance of those learned with task supervised algorithms . TAME evolves morphologies by applying randomly sampled action primitives on a population of agents . This is accomplished using an information-theoretic objective that efficientlyranks agents by their ability to reach diverse states in the environment and the causality of their actions . The code and videos can be found athttps://sites.google.com/view/task-agnostic-evolution.evolution and can be seen in the video below: http://www.mailonline.co.uk/report/tame.org/tAME.com. For more information on TAME please visit www.mailingonline.com .

Author(s) : Donald J. Hejna III, Pieter Abbeel, Lerrel Pinto

Links : PDF - Abstract

Code :

Keywords : tame - task - evolution - agnostic - morphology -

Leave a Reply

Your email address will not be published. Required fields are marked *