{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/unsupervised-learning-of-goal-spaces-for","title":"Unsupervised Learning of Goal Spaces for Intrinsically Motivated Goal Exploration","arxiv_id":"1803.00781","date":"2018-03-02","proceeding":"ICLR 2018 1","authors":["Alexandre Péré","Sébastien Forestier","Olivier Sigaud","Pierre-Yves Oudeyer"],"abstract":"Intrinsically motivated goal exploration algorithms enable machines to\ndiscover repertoires of policies that produce a diversity of effects in complex\nenvironments. These exploration algorithms have been shown to allow real world\nrobots to acquire skills such as tool use in high-dimensional continuous state\nand action spaces. However, they have so far assumed that self-generated goals\nare sampled in a specifically engineered feature space, limiting their\nautonomy. In this work, we propose to use deep representation learning\nalgorithms to learn an adequate goal space. This is a developmental 2-stage\napproach: first, in a perceptual learning stage, deep learning algorithms use\npassive raw sensor observations of world changes to learn a corresponding\nlatent space; then goal exploration happens in a second stage by sampling goals\nin this latent space. We present experiments where a simulated robot arm\ninteracts with an object, and we show that exploration algorithms using such\nlearned representations can match the performance obtained using engineered\nrepresentations.","url_abs":"http://arxiv.org/abs/1803.00781v3","url_pdf":"http://arxiv.org/pdf/1803.00781v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"unsupervised-learning-of-goal-spaces-for","repo_url":"https://github.com/flowersteam/Unsupervised_Goal_Space_Learning","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"AGPL-3.0"}}],"tasks":[{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"representation-learning","task_name":"Representation Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1803.00781","atlas_url":"https://app.syntology.ai/?focus=1803.00781","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}