eprintid: 10083560 rev_number: 28 eprint_status: archive userid: 608 dir: disk0/10/08/35/60 datestamp: 2019-10-18 14:02:21 lastmod: 2021-09-28 22:14:58 status_changed: 2019-10-18 14:03:06 type: proceedings_section metadata_visibility: show creators_name: Sæmundsson, S creators_name: Hofmann, K creators_name: Deisenroth, MP title: Meta reinforcement learning with latent variable Gaussian processes ispublished: pub divisions: UCL divisions: B04 divisions: C05 divisions: F48 note: This version is the version of record. For information on re-use, please refer to the publisher’s terms and conditions. abstract: Learning from small data sets is critical in many practical applications where data collection is time consuming or expensive, e.g., robotics, animal experiments or drug design. Meta learning is one way to increase the data efficiency of learning algorithms by generalizing learned concepts from a set of training tasks to unseen, but related, tasks. Often, this relationship between tasks is hard coded or relies in some other way on human expertise. In this paper, we frame meta learning as a hierarchical latent variable model and infer the relationship between tasks automatically from data. We apply our framework in a modelbased reinforcement learning setting and show that our meta-learning model effectively generalizes to novel tasks by identifying how new tasks relate to prior ones from minimal data. This results in up to a 60% reduction in the average interaction time needed to solve tasks compared to strong baselines. date: 2018-08-06 date_type: published publisher: Association for Uncertainty in Artificial Intelligence (AUAI) official_url: http://auai.org/uai2018/proceedings/papers/235.pdf oa_status: green full_text_type: pub language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 1705333 lyricists_name: Deisenroth, Marc lyricists_id: MDEIS71 actors_name: Nonhebel, Lucinda actors_id: LNONH33 actors_role: owner full_text_status: public series: Uncertainty in Artificial Intelligence (uai) publication: 34th Conference on Uncertainty in Artificial Intelligence 2018, UAI 2018 volume: 34 place_of_pub: Monterey, CA, USA pagerange: 642-652 event_title: 34th Conference on Uncertainty in Artificial Intelligence (uai 2018), 6-10 August 2018, Monterey, CA, USA book_title: Proceedings of 34th Conference on Uncertainty in Artificial Intelligence (uai 2018) editors_name: Elidan, G editors_name: Kersting, K citation: Sæmundsson, S; Hofmann, K; Deisenroth, MP; (2018) Meta reinforcement learning with latent variable Gaussian processes. In: Elidan, G and Kersting, K, (eds.) Proceedings of 34th Conference on Uncertainty in Artificial Intelligence (uai 2018). (pp. pp. 642-652). Association for Uncertainty in Artificial Intelligence (AUAI): Monterey, CA, USA. Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10083560/1/Deisenroth_permitted%20VoR_235.pdf