University of Cambridge > Talks.cam > AI+Pizza > AI+Pizza

AI+Pizza

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Microsoft Research Cambridge Talks Admins.

 Please note, this event may be recorded. Microsoft will own the copyright of any recording and reserves the right to distribute it as required.

Speaker 1: Brooks Paige Title: Predicting Electron Paths Abstract: I will be talking about our recent NIPS submission on modeling chemical reactions by predicting electron paths. Chemical reactions can be described as the stepwise redistribution of electrons in molecules. As such, reactions are often depicted using “arrow-pushing” diagrams which show this movement as a sequence of arrows. We propose an electron path prediction model to learn these sequences directly from data. Instead of predicting product molecules directly from reactant molecules in one shot, learning a model of electron movement has the benefits of (a) being easy for chemists to interpret, (b) incorporating constraints of chemistry, such as balanced atom counts before and after the reaction, and© naturally encoding the sparsity of chemical reactions, which usually involve only a small number of atoms in the reactants. We design a method to extract approximate reaction paths from any dataset of reaction SMILES strings. Furthermore, we show that the model recovers a basic knowledge of chemistry without being explicitly trained to do so. Joint work with John Bradshaw, Matt Kusner, Marwin Segler, and José Miguel Hernández-Lobato.

Speaker 2: Alexander Gaunt Title: Constrained Graph Variational Autoencoders for Molecule Design Abstract: Graphs are ubiquitous data structures for representing interactions between entities. With an emphasis on the use of graphs to represent chemical molecules, 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 more 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.

This talk is part of the AI+Pizza series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2024 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity