• Publications
  • Influence
DeepCoder: Learning to Write Programs
TLDR
The approach is to train a neural network to predict properties of the program that generated the outputs from the inputs to augment search techniques from the programming languages community, including enumerative search and an SMT-based solver.
Constrained Graph Variational Autoencoders for Molecule Design
TLDR
A variational autoencoder model in which both encoder and decoder are graph-structured is proposed and it is shown that by using appropriate shaping of the latent space, this model allows us to design molecules that are (locally) optimal in desired properties.
Learning to Represent Edits
We introduce the problem of learning distributed representations of edits. By combining a "neural editor" with an "edit encoder", our models learn to represent the salient information of an edit and
Deterministic Variational Inference for Robust Bayesian Neural Networks
TLDR
This work introduces a novel deterministic method to approximate moments in neural networks, eliminating gradient variance and introduces a hierarchical prior for parameters and a novel Empirical Bayes procedure for automatically selecting prior variances, and demonstrates good predictive performance over alternative approaches.
Bose-Einstein condensation of atoms in a uniform potential.
TLDR
The Bose-Einstein condensation of an atomic gas in the (quasi)uniform three-dimensional potential of an optical box trap is observed and the critical temperature agrees with the theoretical prediction for a uniform Bose gas.
Generative Code Modeling with Graphs
TLDR
A novel model is presented that uses a graph to represent the intermediate state of the generated output and can generate semantically meaningful expressions, outperforming a range of strong baselines.
TerpreT: A Probabilistic Programming Language for Program Induction
TLDR
The aims are to develop new machine learning approaches based on neural networks and graphical models, and to understand the capabilities of machine learning techniques relative to traditional alternatives, such as those based on constraint solving from the programming languages community.
Differentiable Programs with Neural Libraries
TLDR
A framework for combining differentiable programming languages with neural networks that creates end-to-end trainable systems that learn to write interpretable algorithms with perceptual components and explores the benefits of inductive biases for strong generalization and modularity.
Robust Digital Holography For Ultracold Atom Trapping
TLDR
An improved algorithm for design of arbitrary two-dimensional holographic traps for ultracold atoms is formulated and experimentally demonstrated, which allows for creation of holographic atom traps with a well defined background potential and incorporates full Helmholtz propagation into calculations.
Graph Partition Neural Networks for Semi-Supervised Classification
TLDR
Experimental results indicate that GPNNs are either superior or comparable to state-of-the-art methods on a wide variety of datasets for graph-based semi-supervised classification and can achieve similar performance as standard GNNs with fewer propagation steps.
...
...