File Download
Links for fulltext
(May Require Subscription)
- Scopus: eid_2-s2.0-85064130452
- WOS: WOS:000461852002035
- Find via
Supplementary
- Citations:
- Appears in Collections:
Conference Paper: Constrained graph variational autoencoders for molecule design
Title | Constrained graph variational autoencoders for molecule design |
---|---|
Authors | |
Issue Date | 2018 |
Publisher | Neural Information Processing Systems Foundation |
Citation | 32nd Conference on Neural Information Processing Systems (NeurIPS 2018), Montreal, 2-8 December 2018. In Advances in Neural Information Processing Systems, 2018, v. 2018-December, p. 7795-7804 How to Cite? |
Abstract | Graphs are ubiquitous data structures for representing interactions between entities. With an emphasis on applications in chemistry, we explore the task of learning to generate graphs that conform to a distribution observed in training data. We propose a variational autoencoder model in which both encoder and decoder are graph-structured. Our decoder assumes a sequential ordering of graph extension steps and we discuss and analyze design choices that mitigate the potential downsides of this linearization. Experiments compare our approach with a wide range of baselines on the molecule generation task and show that our method is successful at matching the statistics of the original dataset on semantically important metrics. Furthermore, we show that by using appropriate shaping of the latent space, our model allows us to design molecules that are (locally) optimal in desired properties. |
Persistent Identifier | http://hdl.handle.net/10722/321846 |
ISSN | 2020 SCImago Journal Rankings: 1.399 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Liu, Qi | - |
dc.contributor.author | Allamanis, Miltiadis | - |
dc.contributor.author | Brockschmidt, Marc | - |
dc.contributor.author | Gaunt, Alexander L. | - |
dc.date.accessioned | 2022-11-03T02:21:50Z | - |
dc.date.available | 2022-11-03T02:21:50Z | - |
dc.date.issued | 2018 | - |
dc.identifier.citation | 32nd Conference on Neural Information Processing Systems (NeurIPS 2018), Montreal, 2-8 December 2018. In Advances in Neural Information Processing Systems, 2018, v. 2018-December, p. 7795-7804 | - |
dc.identifier.issn | 1049-5258 | - |
dc.identifier.uri | http://hdl.handle.net/10722/321846 | - |
dc.description.abstract | Graphs are ubiquitous data structures for representing interactions between entities. With an emphasis on applications in chemistry, we explore the task of learning to generate graphs that conform to a distribution observed in training data. We propose a variational autoencoder model in which both encoder and decoder are graph-structured. Our decoder assumes a sequential ordering of graph extension steps and we discuss and analyze design choices that mitigate the potential downsides of this linearization. Experiments compare our approach with a wide range of baselines on the molecule generation task and show that our method is successful at matching the statistics of the original dataset on semantically important metrics. Furthermore, we show that by using appropriate shaping of the latent space, our model allows us to design molecules that are (locally) optimal in desired properties. | - |
dc.language | eng | - |
dc.publisher | Neural Information Processing Systems Foundation | - |
dc.relation.ispartof | Advances in Neural Information Processing Systems | - |
dc.title | Constrained graph variational autoencoders for molecule design | - |
dc.type | Conference_Paper | - |
dc.description.nature | link_to_OA_fulltext | - |
dc.identifier.scopus | eid_2-s2.0-85064130452 | - |
dc.identifier.volume | 2018-December | - |
dc.identifier.spage | 7795 | - |
dc.identifier.epage | 7804 | - |
dc.identifier.isi | WOS:000461852002035 | - |