DeepAI AI Chat
Log In Sign Up

Learning with Molecules beyond Graph Neural Networks

by   Gustav Sourek, et al.

We demonstrate a deep learning framework which is inherently based in the highly expressive language of relational logic, enabling to, among other things, capture arbitrarily complex graph structures. We show how Graph Neural Networks and similar models can be easily covered in the framework by specifying the underlying propagation rules in the relational logic. The declarative nature of the used language then allows to easily modify and extend the propagation schemes into complex structures, such as the molecular rings which we choose for a short demonstration in this paper.


page 1

page 2

page 3

page 4


Beyond Graph Neural Networks with Lifted Relational Neural Networks

We demonstrate a declarative differentiable programming framework based ...

Equivariant Graph Attention Networks for Molecular Property Prediction

Learning and reasoning about 3D molecular structures with varying size i...

Molecular Mechanics-Driven Graph Neural Network with Multiplex Graph for Molecular Structures

The prediction of physicochemical properties from molecular structures i...

Weisfeiler and Leman Go Relational

Knowledge graphs, modeling multi-relational data, improve numerous appli...

Symbolic Relational Deep Reinforcement Learning based on Graph Neural Networks

We present a novel deep reinforcement learning framework for solving rel...

Lifted Relational Neural Networks

We propose a method combining relational-logic representations with neur...

Neural Markov Logic Networks

We introduce Neural Markov Logic Networks (NMLNs), a statistical relatio...