We present a framework for learning hierarchical policies from demonstrations, using sparse natural language annotations to guide the discovery of reusable skills for autonomous decision-making. We formulate a generative model of action sequences in which goals generate sequences of high-level subtask descriptions, and these descriptions generate sequences of low-level actions. We describe how to train this model using primarily unannotated demonstrations by parsing demonstrations into sequences of named high-level sub-tasks, using only a small number of seed annotations to ground language in action. In trained models, natural language commands index a combinatorial library of skills; agents can use these skills to plan by generating high-level instruction sequences tailored to novel goals. We evaluate this approach in the ALFRED household simulation environment, providing natural language annotations for only 10% of demonstrations. It achieves performance comparable state-of-the-art models on ALFRED success rate, outperforming several recent methods with access to ground-truth plans during training and evaluation.
Pratyusha Sharma, Antonio Torralba, and Jacob Andreas. 2022.Skill Induction and Planning with Latent Language. InProceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1713–1726, Dublin, Ireland. Association for Computational Linguistics.
@inproceedings{sharma-etal-2022-skill, title = "Skill Induction and Planning with Latent Language", author = "Sharma, Pratyusha and Torralba, Antonio and Andreas, Jacob", editor = "Muresan, Smaranda and Nakov, Preslav and Villavicencio, Aline", booktitle = "Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)", month = may, year = "2022", address = "Dublin, Ireland", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2022.acl-long.120/", doi = "10.18653/v1/2022.acl-long.120", pages = "1713--1726", abstract = "We present a framework for learning hierarchical policies from demonstrations, using sparse natural language annotations to guide the discovery of reusable skills for autonomous decision-making. We formulate a generative model of action sequences in which goals generate sequences of high-level subtask descriptions, and these descriptions generate sequences of low-level actions. We describe how to train this model using primarily unannotated demonstrations by parsing demonstrations into sequences of named high-level sub-tasks, using only a small number of seed annotations to ground language in action. In trained models, natural language commands index a combinatorial library of skills; agents can use these skills to plan by generating high-level instruction sequences tailored to novel goals. We evaluate this approach in the ALFRED household simulation environment, providing natural language annotations for only 10{\%} of demonstrations. It achieves performance comparable state-of-the-art models on ALFRED success rate, outperforming several recent methods with access to ground-truth plans during training and evaluation."}
<?xml version="1.0" encoding="UTF-8"?><modsCollection xmlns="http://www.loc.gov/mods/v3"><mods ID="sharma-etal-2022-skill"> <titleInfo> <title>Skill Induction and Planning with Latent Language</title> </titleInfo> <name type="personal"> <namePart type="given">Pratyusha</namePart> <namePart type="family">Sharma</namePart> <role> <roleTerm authority="marcrelator" type="text">author</roleTerm> </role> </name> <name type="personal"> <namePart type="given">Antonio</namePart> <namePart type="family">Torralba</namePart> <role> <roleTerm authority="marcrelator" type="text">author</roleTerm> </role> </name> <name type="personal"> <namePart type="given">Jacob</namePart> <namePart type="family">Andreas</namePart> <role> <roleTerm authority="marcrelator" type="text">author</roleTerm> </role> </name> <originInfo> <dateIssued>2022-05</dateIssued> </originInfo> <typeOfResource>text</typeOfResource> <relatedItem type="host"> <titleInfo> <title>Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title> </titleInfo> <name type="personal"> <namePart type="given">Smaranda</namePart> <namePart type="family">Muresan</namePart> <role> <roleTerm authority="marcrelator" type="text">editor</roleTerm> </role> </name> <name type="personal"> <namePart type="given">Preslav</namePart> <namePart type="family">Nakov</namePart> <role> <roleTerm authority="marcrelator" type="text">editor</roleTerm> </role> </name> <name type="personal"> <namePart type="given">Aline</namePart> <namePart type="family">Villavicencio</namePart> <role> <roleTerm authority="marcrelator" type="text">editor</roleTerm> </role> </name> <originInfo> <publisher>Association for Computational Linguistics</publisher> <place> <placeTerm type="text">Dublin, Ireland</placeTerm> </place> </originInfo> <genre authority="marcgt">conference publication</genre> </relatedItem> <abstract>We present a framework for learning hierarchical policies from demonstrations, using sparse natural language annotations to guide the discovery of reusable skills for autonomous decision-making. We formulate a generative model of action sequences in which goals generate sequences of high-level subtask descriptions, and these descriptions generate sequences of low-level actions. We describe how to train this model using primarily unannotated demonstrations by parsing demonstrations into sequences of named high-level sub-tasks, using only a small number of seed annotations to ground language in action. In trained models, natural language commands index a combinatorial library of skills; agents can use these skills to plan by generating high-level instruction sequences tailored to novel goals. We evaluate this approach in the ALFRED household simulation environment, providing natural language annotations for only 10% of demonstrations. It achieves performance comparable state-of-the-art models on ALFRED success rate, outperforming several recent methods with access to ground-truth plans during training and evaluation.</abstract> <identifier type="citekey">sharma-etal-2022-skill</identifier> <identifier type="doi">10.18653/v1/2022.acl-long.120</identifier> <location> <url>https://aclanthology.org/2022.acl-long.120/</url> </location> <part> <date>2022-05</date> <extent unit="page"> <start>1713</start> <end>1726</end> </extent> </part></mods></modsCollection>
%0 Conference Proceedings%T Skill Induction and Planning with Latent Language%A Sharma, Pratyusha%A Torralba, Antonio%A Andreas, Jacob%Y Muresan, Smaranda%Y Nakov, Preslav%Y Villavicencio, Aline%S Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)%D 2022%8 May%I Association for Computational Linguistics%C Dublin, Ireland%F sharma-etal-2022-skill%X We present a framework for learning hierarchical policies from demonstrations, using sparse natural language annotations to guide the discovery of reusable skills for autonomous decision-making. We formulate a generative model of action sequences in which goals generate sequences of high-level subtask descriptions, and these descriptions generate sequences of low-level actions. We describe how to train this model using primarily unannotated demonstrations by parsing demonstrations into sequences of named high-level sub-tasks, using only a small number of seed annotations to ground language in action. In trained models, natural language commands index a combinatorial library of skills; agents can use these skills to plan by generating high-level instruction sequences tailored to novel goals. We evaluate this approach in the ALFRED household simulation environment, providing natural language annotations for only 10% of demonstrations. It achieves performance comparable state-of-the-art models on ALFRED success rate, outperforming several recent methods with access to ground-truth plans during training and evaluation.%R 10.18653/v1/2022.acl-long.120%U https://aclanthology.org/2022.acl-long.120/%U https://doi.org/10.18653/v1/2022.acl-long.120%P 1713-1726
Pratyusha Sharma, Antonio Torralba, and Jacob Andreas. 2022.Skill Induction and Planning with Latent Language. InProceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1713–1726, Dublin, Ireland. Association for Computational Linguistics.