A Deep Generative Model for Molecular Graphs by Niloy Ganguly

A Deep Generative Model for Molecular Graphs by Niloy Ganguly

International Centre for Theoretical Sciences via YouTube Direct link

Start

1 of 25

1 of 25

Start

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

A Deep Generative Model for Molecular Graphs by Niloy Ganguly

Automatically move to the next video in the Classroom when playback concludes

  1. 1 Start
  2. 2 Designing Random Graph Models using Variational Autoencoders...
  3. 3 Discovering new, plausible drug-like molecules
  4. 4 Generative models for molecule design
  5. 5 Limitations of current models
  6. 6 NeVAE: A variational autoencoder for graphs
  7. 7 The probabilistic encoder
  8. 8 Encoder has desirable properties
  9. 9 The probabilistic decoder
  10. 10 Decoder guarantees structural properties
  11. 11 Training is permutation invariant
  12. 12 Training is efficient
  13. 13 Experimental setup
  14. 14 Smooth, meaningful space of molecules
  15. 15 Quantitative evaluation metrics
  16. 16 Competing methods
  17. 17 Validity of the discovered molecules
  18. 18 Novelty of the discovered molecules
  19. 19 Uniqueness of the discovered molecules
  20. 20 Predicting & optimizing for molecule properties
  21. 21 Property prediction Sparse Gaussian Process
  22. 22 Property maximization Bayesian Optimization
  23. 23 Conclusions
  24. 24 Thanks!
  25. 25 Q&A

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.