eprintid: 10088322 rev_number: 23 eprint_status: archive userid: 608 dir: disk0/10/08/83/22 datestamp: 2019-12-19 15:20:10 lastmod: 2021-09-26 23:20:07 status_changed: 2019-12-19 15:20:10 type: proceedings_section metadata_visibility: show creators_name: Janz, D creators_name: Van Der Westhuizen, J creators_name: Paige, B creators_name: Kusner, MJ creators_name: Hernández-Lobato, JM title: Learning a generative model for validity in complex discrete structures ispublished: pub divisions: UCL divisions: B04 divisions: C05 divisions: F48 note: This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. abstract: Deep generative models have been successfully used to learn representations for high-dimensional discrete spaces by representing discrete objects as sequences and employing powerful sequence-based deep models. Unfortunately, these sequence-based models often produce invalid sequences: sequences which do not represent any underlying discrete structure; invalid sequences hinder the utility of such models. As a step towards solving this problem, we propose to learn a deep recurrent validator model, which can estimate whether a partial sequence can function as the beginning of a full, valid sequence. This validator provides insight as to how individual sequence elements influence the validity of the overall sequence, and can be used to constrain sequence based models to generate valid sequences – and thus faithfully model discrete objects. Our approach is inspired by reinforcement learning, where an oracle which can evaluate validity of complete sequences provides a sparse reward signal. We demonstrate its effectiveness as a generative model of Python 3 source code for mathematical expressions, and in improving the ability of a variational autoencoder trained on SMILES strings to decode valid molecular structures. date: 2018-05-03 date_type: published publisher: International Conference on Learning Representations (ICLR) official_url: https://iclr.cc/Conferences/2018/Schedule?showEvent=29 oa_status: green full_text_type: other language: eng primo: open primo_central: open_green article_type_text: Conference Proceeding verified: verified_manual elements_id: 1687067 lyricists_name: Kusner, Matthew lyricists_name: Paige, Timothy lyricists_id: MKKUS92 lyricists_id: BPPAI30 actors_name: Kusner, Matthew actors_id: MKKUS92 actors_role: owner full_text_status: public series: International Conference on Learning Representations (ICLR) publication: 6th International Conference on Learning Representations, ICLR 2018 - Conference Track Proceedings volume: 6 place_of_pub: Vancouver, Canada event_title: Sixth International Conference on Learning Representations (ICLR 2018) book_title: Proceedings of the Sixth International Conference on Learning Representations (ICLR 2018) citation: Janz, D; Van Der Westhuizen, J; Paige, B; Kusner, MJ; Hernández-Lobato, JM; (2018) Learning a generative model for validity in complex discrete structures. In: Proceedings of the Sixth International Conference on Learning Representations (ICLR 2018). International Conference on Learning Representations (ICLR): Vancouver, Canada. Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10088322/1/1712.01664v4.pdf