PulseAugur
EN
LIVE 23:32:06

New quantum and classical models achieve high accuracy in molecular property prediction with few parameters

By PulseAugur Editorial · [2 sources] ·

Researchers have developed two new architectures, Iso-QGNN and Iso-CGNN, for molecular property prediction in low-data quantum chemistry settings. These models leverage a topology-aligned inductive bias, mirroring the molecular bond graph to enhance parameter efficiency. When benchmarked on the QM9 dataset for HOMO-LUMO and dipole moment classification, the Iso-CGNN achieved an AUC of 0.91 on the gap task, while the Iso-QGNN reached 0.88, both with only 64 trainable parameters. The findings suggest this bias is crucial for parameter efficiency and offers a basis for matched-baseline benchmarking in quantum machine learning. AI

IMPACT Introduces parameter-efficient architectures for quantum chemistry, potentially accelerating drug discovery and materials science research.

RANK_REASON The cluster contains an arXiv preprint detailing new research on machine learning architectures for molecular property prediction.

Read on arXiv cs.LG →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

New quantum and classical models achieve high accuracy in molecular property prediction with few parameters

COVERAGE [2]

  1. arXiv cs.LG TIER_1 English(EN) · James T. Pegg, Hubert Okadome Valencia, Ronin Wu ·

    Implementations of Quantum and Classical Topology-Aligned Architectures for Molecular Property Prediction

    arXiv:2607.13737v1 Announce Type: new Abstract: For low-data and resource-constrained regimes typical of quantum chemistry, parameter-efficient learning is a key objective. Here, we propose a topology-aligned inductive bias in which the model architecture mirrors the molecular bo…

  2. arXiv cs.LG TIER_1 English(EN) · Ronin Wu ·

    Implementations of Quantum and Classical Topology-Aligned Architectures for Molecular Property Prediction

    For low-data and resource-constrained regimes typical of quantum chemistry, parameter-efficient learning is a key objective. Here, we propose a topology-aligned inductive bias in which the model architecture mirrors the molecular bond graph: atoms map to a fixed register of compu…

RELATED ENTITIES

RELATED TOPICS