Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Mario Geiger

Mario Geiger

An end-to-end SE(3)-equivariant segmentation network

Mar 02, 2023
Ivan Diaz, Mario Geiger, Richard Iain McKinley

Figure 1 for An end-to-end SE(3)-equivariant segmentation network

Figure 2 for An end-to-end SE(3)-equivariant segmentation network

Figure 3 for An end-to-end SE(3)-equivariant segmentation network

Figure 4 for An end-to-end SE(3)-equivariant segmentation network

* 19 pages, 10 figures, submitted to the Journal of Machine Learning for Biomedical Imaging

Access Paper or Ask Questions

Dissecting the Effects of SGD Noise in Distinct Regimes of Deep Learning

Jan 31, 2023
Antonio Sclocchi, Mario Geiger, Matthieu Wyart

Figure 1 for Dissecting the Effects of SGD Noise in Distinct Regimes of Deep Learning

Figure 2 for Dissecting the Effects of SGD Noise in Distinct Regimes of Deep Learning

Figure 3 for Dissecting the Effects of SGD Noise in Distinct Regimes of Deep Learning

Figure 4 for Dissecting the Effects of SGD Noise in Distinct Regimes of Deep Learning

* 18 pages, 14 figures

Access Paper or Ask Questions

e3nn: Euclidean Neural Networks

Jul 18, 2022
Mario Geiger, Tess Smidt

Figure 1 for e3nn: Euclidean Neural Networks

Figure 2 for e3nn: Euclidean Neural Networks

Figure 3 for e3nn: Euclidean Neural Networks

Figure 4 for e3nn: Euclidean Neural Networks

* draft

Access Paper or Ask Questions

Relevance of Rotationally Equivariant Convolutions for Predicting Molecular Properties

Aug 22, 2020
Benjamin Kurt Miller, Mario Geiger, Tess E. Smidt, Frank Noé

Figure 1 for Relevance of Rotationally Equivariant Convolutions for Predicting Molecular Properties

Figure 2 for Relevance of Rotationally Equivariant Convolutions for Predicting Molecular Properties

Figure 3 for Relevance of Rotationally Equivariant Convolutions for Predicting Molecular Properties

Figure 4 for Relevance of Rotationally Equivariant Convolutions for Predicting Molecular Properties

* NeurIPS 2020 Machine Learning for Molecules

Access Paper or Ask Questions