Methods › General › Activation Functions › Shifted Softplus
Shifted Softplus
Introduced by Kristof T. Schütt et al. in SchNet: A continuous-filter convolutional neural network for modeling quantum interactions
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
Shifted Softplus is an activation function ssp(x) = ln( 0.5 eˣ + 0.5 ), which SchNet employs as non-linearity throughout the network in order to obtain a smooth potential energy surface. The shifting ensures that ssp(0) = 0 and improves the convergence of the network. This activation function shows similarity to ELUs, while having infinite order of continuity.
Papers archive 2025-07-28
20 shown of 20, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.
-
Uncertainty Quantification in Graph Neural Networks with Shallow Ensembles 17 Apr 2025 · 0 repositories · arXiv:2504.12627
-
Machine learning surrogate models of many-body dispersion interactions in polymer melts 19 Mar 2025 · 0 repositories · arXiv:2503.15149
-
Machine Learned Force Fields: Fundamentals, its reach, and challenges 7 Mar 2025 · 0 repositories · arXiv:2503.05845
-
OpenQDC: Open Quantum Data Commons 29 Nov 2024 · 1 repository · arXiv:2411.19629
-
SynCoTrain: A Dual Classifier PU-learning Framework for Synthesizability Prediction 18 Nov 2024 · 0 repositories · arXiv:2411.12011
-
Distribution Learning for Molecular Regression 30 Jul 2024 · 0 repositories · arXiv:2407.20475
-
Lightweight Geometric Deep Learning for Molecular Modelling in Catalyst Discovery 5 Apr 2024 · 0 repositories · arXiv:2404.10003
-
On the importance of catalyst-adsorbate 3D interactions for relaxed energy predictions 10 Oct 2023 · 0 repositories · arXiv:2310.06682
-
End-to-end AI framework for interpretable prediction of molecular and crystal properties 21 Dec 2022 · 1 repository · arXiv:2212.11317
-
Machine Learning for Screening Large Organic Molecules 23 Nov 2022 · 0 repositories · arXiv:2211.15415
-
Developing Machine-Learned Potentials for Coarse-Grained Molecular Simulations: Challenges and Pitfalls 26 Sep 2022 · 0 repositories · arXiv:2209.12948
-
Predicting Aqueous Solubility of Organic Molecules Using Deep Learning Models with Varied Molecular Representations 26 May 2021 · 1 repository · arXiv:2105.12638
-
The Open Catalyst 2020 (OC20) Dataset and Community Challenges 20 Oct 2020 · 5 repositories · arXiv:2010.09990Syntology ran 0 of 11 samples · 11 unverified · 11 pointer-only (licence)
-
Heterogeneous Molecular Graph Neural Networks for Predicting Molecule Properties 26 Sep 2020 · 1 repository · arXiv:2009.12710Syntology ran 3 of 3 samples · 0 unverified · 3 pointer-only (licence)
-
A deep neural network for molecular wave functions in quasi-atomic minimal basis representation 11 May 2020 · 0 repositories · arXiv:2005.06979
-
Combining SchNet and SHARC: The SchNarc machine learning approach for excited-state dynamics 17 Feb 2020 · 1 repository · arXiv:2002.07264Syntology ran 0 of 8 samples · 8 unverified
-
Learning representations of molecules and materials with atomistic neural networks 11 Dec 2018 · 0 repositories · arXiv:1812.04690
-
Analysis of Atomistic Representations Using Weighted Skip-Connections 23 Oct 2018 · 0 repositories · arXiv:1810.09751
-
Neural Message Passing with Edge Updates for Predicting Properties of Molecules and Materials 8 Jun 2018 · 5 repositories · arXiv:1806.03146Syntology ran 4 of 22 samples · 18 unverified · 4 pointer-only (licence)
-
SchNet: A continuous-filter convolutional neural network for modeling quantum interactions 26 Jun 2017 · 5 repositories · arXiv:1706.08566
Tasks archive 2025-07-28
19 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections