?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=Learning+a+generative+model+for+validity+in+complex+discrete+structures&rft.creator=Janz%2C+D&rft.creator=Van+Der+Westhuizen%2C+J&rft.creator=Paige%2C+B&rft.creator=Kusner%2C+MJ&rft.creator=Hern%C3%A1ndez-Lobato%2C+JM&rft.description=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%2C+these+sequence-based+models+often+produce+invalid+sequences%3A+sequences+which+do+not+represent+any+underlying+discrete+structure%3B+invalid+sequences+hinder+the+utility+of+such+models.+As+a+step+towards+solving+this+problem%2C+we+propose+to+learn+a+deep+recurrent+validator+model%2C+which+can+estimate+whether+a+partial+sequence+can+function+as+the+beginning+of+a+full%2C+valid+sequence.+This+validator+provides+insight+as+to+how+individual+sequence+elements+influence+the+validity+of+the+overall+sequence%2C+and+can+be+used+to+constrain+sequence+based+models+to+generate+valid+sequences+%E2%80%93+and+thus+faithfully+model+discrete+objects.+Our+approach+is+inspired+by+reinforcement+learning%2C+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%2C+and+in+improving+the+ability+of+a+variational+autoencoder+trained+on+SMILES+strings+to+decode+valid+molecular+structures.&rft.publisher=International+Conference+on+Learning+Representations+(ICLR)&rft.date=2018-05-03&rft.type=Proceedings+paper&rft.language=eng&rft.source=+++++In%3A++Proceedings+of+the+Sixth+International+Conference+on+Learning+Representations+(ICLR+2018).++++International+Conference+on+Learning+Representations+(ICLR)%3A+Vancouver%2C+Canada.+(2018)+++++&rft.format=text&rft.identifier=https%3A%2F%2Fdiscovery.ucl.ac.uk%2Fid%2Feprint%2F10088322%2F1%2F1712.01664v4.pdf&rft.identifier=https%3A%2F%2Fdiscovery.ucl.ac.uk%2Fid%2Feprint%2F10088322%2F&rft.rights=open