Research map / Graph learning / GNN architectures
A Function-Centric Graph Neural Network Approach for Predicting Electron Densities
Manuel V. Klockow, Marc K. Ickler, Peter Lippmann, Fred A. Hamprecht
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add smooth absolute value (without radial correction) (ablation) | BOA small · QM9 VASP | NMAE [%]: 0.204 -> 0.167 | Improved |
| add smooth absolute value of one function in each pair (with radial correction) (ablation) | BOA small · QM9 VASP | NMAE [%]: 0.1423 -> 0.1381 | Improved |
| add radial correction (ablation) | BOA small · QM9 VASP | NMAE [%]: 0.204 -> 0.1423 | Improved |
| use quadratic instead of linear density expansion (ablation) | BOA small · QM9 VASP | NMAE [%]: 0.2716 -> 0.1381 | Improved |
| use larger basis set def2-QZVPPD instead of def2-TZVP (ablation) | BOA small · QM9 VASP | NMAE [%]: 0.1504 -> 0.1381 | Improved |
| use larger basis set def2-TZVP instead of def2-SVP (ablation) | BOA small · QM9 VASP | NMAE [%]: 0.194 -> 0.1504 | Improved |
| use larger cutoffs (rmp=8, re=4) (ablation) | BOA small · QM9 VASP | NMAE [%]: 0.1381 -> 0.1343 | Improved |
| use larger model (BOA large vs small) | BOA · QM9 PySCF | NMAE [%]: 0.13 -> 0.116 | Improved |
| use smaller cutoffs (rmp=3, re=2) for generalization to larger molecules (ablation) | BOA · QMugs | NMAE [%] | Improved |
| use def2-TZVP basis to cut inference time (ablation) | BOA · QM9 VASP | time per molecule [s]: 1.27 -> 0.64 | Improved |
| initialize KS-DFT with BOA density instead of atomic superposition | KS-DFT · QM9 PySCF | mean SCF iterations: 15.7 -> 10.2 | Improved |
Closest papers by meaning
- Learning the Electronic Hamiltonian of Large Atomic Structures (ICML 2025 · similarity 0.89)
- A Recipe for Charge Density Prediction (NeurIPS 2024 · similarity 0.88)
- Gaussian Plane-Wave Neural Operator for Electron Density Estimation (ICML 2024 · similarity 0.87)
- Learning Equivariant Non-Local Electron Density Functionals (ICLR 2025 · similarity 0.87)
- ELECTRA: A Cartesian Network for 3D Charge Density Prediction with Floating Orbitals (NeurIPS 2025 · similarity 0.86)
- Efficient and Scalable Density Functional Theory Hamiltonian Prediction through Adaptive Sparsity (ICML 2025 · similarity 0.85)
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.
