# Multitask methods for predicting molecular properties from heterogeneous data

arXiv: https://arxiv.org/abs/2401.17898 · HTML: https://mlsanity.com/paper/2401.17898/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 66 references that are in the map.
- First posted on arXiv 31 Jan 2024.

## Builds on (most cited first, 12 of 12)

- [The Open Catalyst 2020 (OC20) Dataset and Community Challenges](https://mlsanity.com/paper/2010.09990/) (2020)
- [Deep learning: a statistical viewpoint](https://mlsanity.com/paper/2103.09177/) (2021)
- [When Gaussian Process Meets Big Data: A Review of Scalable GPs](https://mlsanity.com/paper/1807.01065/) (2018)
- [Interactive Supercomputing on 40,000 Cores for Machine Learning and Data Analysis](https://mlsanity.com/paper/1807.07814/) (2018)
- [PAC-Bayes Compression Bounds So Tight That They Can Explain Generalization](https://mlsanity.com/paper/2211.13609/) (2022)
- [DScribe: Library of Descriptors for Machine Learning in Materials Science](https://mlsanity.com/paper/1904.08875/) (2019)
- [Thoughts on Massively Scalable Gaussian Processes](https://mlsanity.com/paper/1511.01870/) (2015)
- [AiiDA 1.0, a scalable computational infrastructure for automated reproducible workflows and data provenance](https://mlsanity.com/paper/2003.12476/) (2020)
- [Locally induced Gaussian processes for large-scale simulation experiments](https://mlsanity.com/paper/2008.12857/) (2020)
- [Transfer learning for chemically accurate interatomic neural network potentials](https://mlsanity.com/paper/2212.03916/) (2022)
- [Machine learning to tame divergent density functional approximations: a new path to consensus materials design principles](https://mlsanity.com/paper/2106.13109/) (2021)
- [Multi-Fidelity Machine Learning for Excited State Energies of Molecules](https://mlsanity.com/paper/2305.11292/) (2023)

## Built on by (most cited first, 8 of 8)

- [QeMFi: A Multifidelity Dataset of Quantum Chemical Properties of Diverse Molecules](https://mlsanity.com/paper/2406.14149/) (2024)
- [Assessing Non-Nested Configurations of Multifidelity Machine Learning for Quantum-Chemical Properties](https://mlsanity.com/paper/2407.17087/) (2024)
- [Investigating Data Hierarchies in Multifidelity Machine Learning for Excitation Energies](https://mlsanity.com/paper/2410.11392/) (2024)
- [Data Fusion of Deep Learned Molecular Embeddings for Property Prediction](https://mlsanity.com/paper/2504.07297/) (2025)
- [LFaB: Low fidelity as Bias for Active Learning in the chemical configuration space](https://mlsanity.com/paper/2508.15577/) (2025)
- [Quantum-Enhanced Multi-Task Learning with Learnable Weighting for Pharmacokinetic and Toxicity Prediction](https://mlsanity.com/paper/2509.04601/) (2025)
- [Benchmarking Data Efficiency in Delta-ML and Multifidelity Models for Quantum Chemistry](https://mlsanity.com/paper/2410.11391/) (2024)
- [Multivariate conformal uncertainty propagation in multitask atomistic simulation: Successes and pitfalls](https://mlsanity.com/paper/2609.31384/) (2026)

## Abstract (arXiv)

Data generation remains a bottleneck in training surrogate models to predict molecular properties. We demonstrate that multitask Gaussian process regression overcomes this limitation by leveraging both expensive and cheap data sources. In particular, we consider training sets constructed from coupled-cluster (CC) and density functional theory (DFT) data. We report that multitask surrogates can predict at CC-level accuracy with a reduction to data generation cost by over an order of magnitude. Of note, our approach allows the training set to include DFT data generated by a heterogeneous mix of exchange-correlation functionals without imposing any artificial hierarchy on functional accuracy. More generally, the multitask framework can accommodate a wider range of training set structures -- including full disparity between the different levels of fidelity -- than existing kernel approaches based on Delta-learning, though we show that the accuracy of the two approaches can be similar. Consequently, multitask regression can be a tool for reducing data generation costs even further by opportunistically exploiting existing data sources.