Learning how to actively learn: a deep imitation learning approach

    Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

    Abstract

    Heuristic-based active learning (AL) methods are limited when the data distribution of the underlying learning problems vary. We introduce a method
    that learns an AL policy using imitation learning (IL). Our IL-based approach
    makes use of an efficient and effective algorithmic expert, which provides the
    policy learner with good actions in the encountered AL situations. The AL strategy is then learned with a feedforward network, mapping situations to most
    informative query datapoints. We evaluate our method on two different tasks: text classification and named entity recognition. Experimental results show that our IL-based AL strategy is more effective than strong previous methods using
    heuristics and reinforcement learning.
    Original languageEnglish
    Title of host publicationACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics
    Subtitle of host publicationProceedings of the Conference, Vol. 1 (Long Papers)
    EditorsIryna Gurevych, Yusuke Miyao
    Place of PublicationStroudsburg PA USA
    PublisherAssociation for Computational Linguistics (ACL)
    Pages1874-1883
    Number of pages10
    ISBN (Print)9781948087322
    Publication statusPublished - 2018
    EventAnnual Meeting of the Association for Computational Linguistics 2018 - Melbourne, Australia
    Duration: 15 Jul 201820 Jul 2018
    Conference number: 56th
    https://aclanthology.info/events/acl-2018

    Conference

    ConferenceAnnual Meeting of the Association for Computational Linguistics 2018
    Abbreviated titleACL 2018
    CountryAustralia
    CityMelbourne
    Period15/07/1820/07/18
    Internet address

    Cite this

    Liu, M., Buntine, W., & Haffari, G. (2018). Learning how to actively learn: a deep imitation learning approach. In I. Gurevych, & Y. Miyao (Eds.), ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics: Proceedings of the Conference, Vol. 1 (Long Papers) (pp. 1874-1883). Stroudsburg PA USA: Association for Computational Linguistics (ACL).
    Liu, Ming ; Buntine, Wray ; Haffari, Gholamreza. / Learning how to actively learn : a deep imitation learning approach. ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics: Proceedings of the Conference, Vol. 1 (Long Papers). editor / Iryna Gurevych ; Yusuke Miyao. Stroudsburg PA USA : Association for Computational Linguistics (ACL), 2018. pp. 1874-1883
    @inproceedings{c231a236aabb4788a87c4f18d7ef3e6b,
    title = "Learning how to actively learn: a deep imitation learning approach",
    abstract = "Heuristic-based active learning (AL) methods are limited when the data distribution of the underlying learning problems vary. We introduce a methodthat learns an AL policy using imitation learning (IL). Our IL-based approachmakes use of an efficient and effective algorithmic expert, which provides thepolicy learner with good actions in the encountered AL situations. The AL strategy is then learned with a feedforward network, mapping situations to mostinformative query datapoints. We evaluate our method on two different tasks: text classification and named entity recognition. Experimental results show that our IL-based AL strategy is more effective than strong previous methods usingheuristics and reinforcement learning.",
    author = "Ming Liu and Wray Buntine and Gholamreza Haffari",
    year = "2018",
    language = "English",
    isbn = "9781948087322",
    pages = "1874--1883",
    editor = "Iryna Gurevych and Yusuke Miyao",
    booktitle = "ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics",
    publisher = "Association for Computational Linguistics (ACL)",

    }

    Liu, M, Buntine, W & Haffari, G 2018, Learning how to actively learn: a deep imitation learning approach. in I Gurevych & Y Miyao (eds), ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics: Proceedings of the Conference, Vol. 1 (Long Papers). Association for Computational Linguistics (ACL), Stroudsburg PA USA, pp. 1874-1883, Annual Meeting of the Association for Computational Linguistics 2018, Melbourne, Australia, 15/07/18.

    Learning how to actively learn : a deep imitation learning approach. / Liu, Ming; Buntine, Wray; Haffari, Gholamreza.

    ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics: Proceedings of the Conference, Vol. 1 (Long Papers). ed. / Iryna Gurevych; Yusuke Miyao. Stroudsburg PA USA : Association for Computational Linguistics (ACL), 2018. p. 1874-1883.

    Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

    TY - GEN

    T1 - Learning how to actively learn

    T2 - a deep imitation learning approach

    AU - Liu, Ming

    AU - Buntine, Wray

    AU - Haffari, Gholamreza

    PY - 2018

    Y1 - 2018

    N2 - Heuristic-based active learning (AL) methods are limited when the data distribution of the underlying learning problems vary. We introduce a methodthat learns an AL policy using imitation learning (IL). Our IL-based approachmakes use of an efficient and effective algorithmic expert, which provides thepolicy learner with good actions in the encountered AL situations. The AL strategy is then learned with a feedforward network, mapping situations to mostinformative query datapoints. We evaluate our method on two different tasks: text classification and named entity recognition. Experimental results show that our IL-based AL strategy is more effective than strong previous methods usingheuristics and reinforcement learning.

    AB - Heuristic-based active learning (AL) methods are limited when the data distribution of the underlying learning problems vary. We introduce a methodthat learns an AL policy using imitation learning (IL). Our IL-based approachmakes use of an efficient and effective algorithmic expert, which provides thepolicy learner with good actions in the encountered AL situations. The AL strategy is then learned with a feedforward network, mapping situations to mostinformative query datapoints. We evaluate our method on two different tasks: text classification and named entity recognition. Experimental results show that our IL-based AL strategy is more effective than strong previous methods usingheuristics and reinforcement learning.

    M3 - Conference Paper

    SN - 9781948087322

    SP - 1874

    EP - 1883

    BT - ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics

    A2 - Gurevych, Iryna

    A2 - Miyao, Yusuke

    PB - Association for Computational Linguistics (ACL)

    CY - Stroudsburg PA USA

    ER -

    Liu M, Buntine W, Haffari G. Learning how to actively learn: a deep imitation learning approach. In Gurevych I, Miyao Y, editors, ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics: Proceedings of the Conference, Vol. 1 (Long Papers). Stroudsburg PA USA: Association for Computational Linguistics (ACL). 2018. p. 1874-1883