Deep Learning Foundation Models from Classical Molecular Descriptors
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908824010489856 |
|---|---|
| author | Burns, Jackson W. Zalte, Akshat Shirish Abreu, Charlles R. A. Sieg, Jochen Feldmann, Christian Mathea, Miriam Green, William H. |
| author_facet | Burns, Jackson W. Zalte, Akshat Shirish Abreu, Charlles R. A. Sieg, Jochen Feldmann, Christian Mathea, Miriam Green, William H. |
| contents | Fast and accurate data-driven prediction of molecular properties is pivotal to scientific advancements across myriad chemical domains. Deep learning methods have recently garnered much attention, despite their inability to outperform classical machine learning methods when tested on practical, real-world benchmarks with limited training data. This study seeks to bridge this gap with CheMeleon, a O(10M) parameter foundation model that enables directed message-passing neural networks to finally exceed the performance of classical methods. Evaluated on 58 benchmark datasets from Polaris and MoleculeACE, CheMeleon achieves a win rate of 75% on Polaris tasks, outperforming baselines like Random Forest (68%), fastprop (36%), and Chemprop (32%), and a 97% win rate on MoleculeACE assays, surpassing Random Forest (50%) and other foundation models. Unlike conventional pre-training approaches that rely on noisy experimental data or biased quantum mechanical simulations, CheMeleon utilizes low-noise molecular descriptors to learn rich and highly transferable molecular representations, suggesting a new avenue for foundation model pre-training. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_15792 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Deep Learning Foundation Models from Classical Molecular Descriptors Burns, Jackson W. Zalte, Akshat Shirish Abreu, Charlles R. A. Sieg, Jochen Feldmann, Christian Mathea, Miriam Green, William H. Machine Learning Chemical Physics Fast and accurate data-driven prediction of molecular properties is pivotal to scientific advancements across myriad chemical domains. Deep learning methods have recently garnered much attention, despite their inability to outperform classical machine learning methods when tested on practical, real-world benchmarks with limited training data. This study seeks to bridge this gap with CheMeleon, a O(10M) parameter foundation model that enables directed message-passing neural networks to finally exceed the performance of classical methods. Evaluated on 58 benchmark datasets from Polaris and MoleculeACE, CheMeleon achieves a win rate of 75% on Polaris tasks, outperforming baselines like Random Forest (68%), fastprop (36%), and Chemprop (32%), and a 97% win rate on MoleculeACE assays, surpassing Random Forest (50%) and other foundation models. Unlike conventional pre-training approaches that rely on noisy experimental data or biased quantum mechanical simulations, CheMeleon utilizes low-noise molecular descriptors to learn rich and highly transferable molecular representations, suggesting a new avenue for foundation model pre-training. |
| title | Deep Learning Foundation Models from Classical Molecular Descriptors |
| topic | Machine Learning Chemical Physics |
| url | https://arxiv.org/abs/2506.15792 |