AI Paper Atlas

Research map / Graph learning / GNN architectures

Transfer learning for atomistic simulations using GNNs and kernel mean embeddings

John Isak Texas Falk, Luigi Bonati, Pietro Novelli, Michele Parrinello, Massimiliano Pontil

NeurIPS 2023 · Graph learning › GNN architectures · approach: Physics-inspired GNNs · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
MEKRR-(SCN) instead of SchNet-FTMEKRR · Fe/N2 D2RMSE: 2.0 -> 0.9Improved
MEKRR-(SchNet) instead of SchNet trained from scratchSchNet · Cu/formateRMSE: 6.0 -> 1.2Improved
MEKRR-(SCN) instead of GAPGAP · Fe/N2 D4RMSE: 4.9 -> 2.7Improved
MEKRR-(SchNet) instead of SchNet-FT in transferSchNet-FT · D1->D2RMSE: 17.6 -> 8.0Improved
MEKRR-(SCN) instead of SchNet-FT in transferSchNet-FT · D3->D4RMSE: 116 -> 42Improved
MEKRR-(SchNet) instead of SchNet-FTSchNet-FT · Fe/N2 D4RMSE: 3.2 -> 3.3No clear effect
Gaussian kernel (MEKRR) instead of linear kernel (ablation)KRR alpha=1 · Fe/N2 D1RMSE: 21 -> 0.1Improved
Gaussian kernel (MEKRR) instead of linear kernel (ablation)KRR alpha=0 · Fe/N2 D4RMSE: 146 -> 2.2Improved
species-weighted alpha=1 instead of shared weights alpha=0 (ablation)MEKRR · Fe/N2 D1RMSE: 0.3 -> 0.1Improved
species-weighted alpha=1 instead of shared weights alpha=0 (ablation)MEKRR · Fe/N2 D4RMSE: 2.2 -> 3.3Hurt
deeper representation layer 5 instead of layer 1 (ablation)MEKRR alpha=1 · Fe/N2 D2RMSE: 1.7 -> 0.9Improved
deeper representation layer 5 instead of layer 1 (ablation)MEKRR alpha=0 · Fe/N2 D2RMSE: 1.9 -> 1.4Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy