paper-with-me

Papers

Tensor Frames -- How To Make Any Message Passing Network Equivariant

2024-05-24 · Peter Lippmann, Gerrit Gerhartz, Roman Remme, Fred A. Hamprecht

In many applications of geometric deep learning, the choice of global coordinate frame is arbitrary, and predictions should be independent of the reference frame. In other words, the network should be equivariant with respect to rotations and reflections of the input, i.e., the transformations of O(d). We present a novel framework for building equivariant message passing architectures and modifying existing non-equivariant architectures to be equivariant. Our approach is based on local coordinate frames, between which geometric information is communicated consistently by including tensorial objects in the messages. Our framework can be applied to message passing on geometric data in arbitrary dimensional Euclidean space. While many other approaches for equivariant message passing require specialized building blocks, such as non-standard normalization layers or non-linearities, our approach can be adapted straightforwardly to any existing architecture without such modifications. We explicitly demonstrate the benefit of O(3)-equivariance for a popular point cloud architecture and produce state-of-the-art results on normal vector regression on point clouds.

📄 PDF Abstract BibTeX arXiv:2405.15389

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Direct Molecular Polarizability Prediction with SO(3) Equivariant Local Frame GNNs

2025-11-10 · Jean Philip Filling, Felix Post, Michael Wand, Denis Andrienko arxiv

We introduce a novel equivariant graph neural network (GNN) architecture designed to predict the tensorial response properties of molecules. Unlike traditional frameworks that focus on regressing scalar quantities and de…

Molecular Property PredictionGraph Neural Network

Lorentz Local Canonicalization: How to Make Any Network Lorentz-Equivariant

2025-05-26 · Jonas Spinner, Luigi Favaro, Peter Lippmann, Sebastian Pitz 외

Lorentz-equivariant neural networks are becoming the leading architectures for high-energy physics. Current implementations rely on specialized layers, limiting architectural choices. We introduce Lorentz Local Canonical…

Data Augmentation

Equivariance by Local Canonicalization: A Matter of Representation

2025-09-30 · Gerrit Gerhartz, Peter Lippmann, Fred A. Hamprecht arxiv

Equivariant neural networks offer strong inductive biases for learning from molecular and geometric data but often rely on specialized, computationally expensive tensor operations. We present a framework to transfers exi…

Learning Local Equivariant Representations for Large-Scale Atomistic Dynamics

2022-04-11 · Albert Musaelian, Simon Batzner, Anders Johansson, Lixin Sun 외

A simultaneously accurate and computationally efficient parametrization of the energy and atomic forces of molecules and materials is a long-standing goal in the natural sciences. In pursuit of this goal, neural message …

Atomic Forces

P-tensors: a General Formalism for Constructing Higher Order Message Passing Networks

2023-06-19 · Tianyi Sun, Andrew Hands, Risi Kondor

Several recent papers have recently shown that higher order graph neural networks can achieve better accuracy than their standard message passing counterparts, especially on highly structured graphs such as molecules. Th…