Graph Deep Learning Lab
We investigate fundamental techniques in Graph Deep Learning, a new framework that combines graph theory and deep neural networks to tackle complex data domains in physical science, natural language processing, computer vision, and combinatorial optimization.
Projects
*
Benchmarking Graph Neural Networks
Identify universal building blocks for robust and scalable GNNs.
Free-hand Sketches
Representation learning for drawings via graphs with geometric and temporal information.
Combinatorial Optimization
Scalable deep learning systems for practical NP-Hard combinatorial problems such as the TSP.
Quantum Chemistry
Chemical synthesis, structure and property prediction using deep neural networks.
Spatial Graph ConvNets
Graph Neural Network architectures for inductive representation learning on arbitrary graphs.
Recent Publications
Quickly discover relevant content by filtering publications.
Learning TSP Requires Rethinking Generalization
End-to-end training of neural network solvers for combinatorial problems such as the Travelling Salesman Problem is intractable and …
Benchmarking Graph Neural Networks
Graph neural networks (GNNs) have become the standard toolkit for analyzing and learning from data on graphs. As the field grows, it …
Multi-Graph Transformer for Free-Hand Sketch Recognition
Learning meaningful representations of free-hand sketches remains a challenging task given the signal sparsity and the high-level …
A Two-Step Graph Convolutional Decoder for Molecule Generation
We propose a simple auto-encoder framework for molecule generation. The molecular graph is first encoded into a continuous latent …
On Learning Paradigms for the Travelling Salesman Problem
We explore the impact of learning paradigms on training deep neural networks for the Travelling Salesman Problem. We design controlled …
Recent Blogposts
Benchmarking Graph Neural Networks
This blog is based on the paper Benchmarking Graph Neural Networks which is a joint work with Chaitanya K. Joshi, Thomas Laurent, …
Transformers are Graph Neural Networks
Engineer friends often ask me: Graph Deep Learning sounds great, but are there any big commercial success stories? Is it being deployed …
Recent Talks
Graph Neural Networks for the Travelling Salesman Problem
The most famous NP-hard combinatorial problem today, the Travelling Salesman Problem, is intractable to solve optimally at large scale. …
Graph Convolutional Neural Networks for Molecule Generation
In this talk, I will discuss a graph convolutional neural network architecture for the molecule generation task. The proposed approach …
Graph Convolutional Neural Networks for Molecule Generation and Travelling Salesman Problem
In this talk, I will discuss how to apply graph convolutional neural networks to quantum chemistry and operational research. The same …
People
Principal Investigator
Xavier Bresson
Assoc. Professor of Computer Science
PhD Students
Vijay Prakash Dwivedi
PhD Student
David Low Jia Wei
PhD Student
Postdoctoral Scholars
Peng Xu
Postdoctoral Scholar
Research Assistants
Chaitanya Joshi
Research Assistant
Victor Getty
Research Assistant
Visitors
