Multi-Fidelity Active Learning with GFlowNets

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hernandez-Garcia, Alex, Saxena, Nikita, Jain, Moksh, Liu, Cheng-Hao, Bengio, Yoshua
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909301695578112
author Hernandez-Garcia, Alex
Saxena, Nikita
Jain, Moksh
Liu, Cheng-Hao
Bengio, Yoshua
author_facet Hernandez-Garcia, Alex
Saxena, Nikita
Jain, Moksh
Liu, Cheng-Hao
Bengio, Yoshua
contents In the last decades, the capacity to generate large amounts of data in science and engineering applications has been growing steadily. Meanwhile, machine learning has progressed to become a suitable tool to process and utilise the available data. Nonetheless, many relevant scientific and engineering problems present challenges where current machine learning methods cannot yet efficiently leverage the available data and resources. For example, in scientific discovery, we are often faced with the problem of exploring very large, structured and high-dimensional spaces. Moreover, the high fidelity, black-box objective function is often very expensive to evaluate. Progress in machine learning methods that can efficiently tackle such challenges would help accelerate currently crucial areas such as drug and materials discovery. In this paper, we propose a multi-fidelity active learning algorithm with GFlowNets as a sampler, to efficiently discover diverse, high-scoring candidates where multiple approximations of the black-box function are available at lower fidelity and cost. Our evaluation on molecular discovery tasks shows that multi-fidelity active learning with GFlowNets can discover high-scoring candidates at a fraction of the budget of its single-fidelity counterpart while maintaining diversity, unlike RL-based alternatives. These results open new avenues for multi-fidelity active learning to accelerate scientific discovery and engineering design.
format Preprint
id arxiv_https___arxiv_org_abs_2306_11715
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Multi-Fidelity Active Learning with GFlowNets
Hernandez-Garcia, Alex
Saxena, Nikita
Jain, Moksh
Liu, Cheng-Hao
Bengio, Yoshua
Machine Learning
Biomolecules
In the last decades, the capacity to generate large amounts of data in science and engineering applications has been growing steadily. Meanwhile, machine learning has progressed to become a suitable tool to process and utilise the available data. Nonetheless, many relevant scientific and engineering problems present challenges where current machine learning methods cannot yet efficiently leverage the available data and resources. For example, in scientific discovery, we are often faced with the problem of exploring very large, structured and high-dimensional spaces. Moreover, the high fidelity, black-box objective function is often very expensive to evaluate. Progress in machine learning methods that can efficiently tackle such challenges would help accelerate currently crucial areas such as drug and materials discovery. In this paper, we propose a multi-fidelity active learning algorithm with GFlowNets as a sampler, to efficiently discover diverse, high-scoring candidates where multiple approximations of the black-box function are available at lower fidelity and cost. Our evaluation on molecular discovery tasks shows that multi-fidelity active learning with GFlowNets can discover high-scoring candidates at a fraction of the budget of its single-fidelity counterpart while maintaining diversity, unlike RL-based alternatives. These results open new avenues for multi-fidelity active learning to accelerate scientific discovery and engineering design.
title Multi-Fidelity Active Learning with GFlowNets
topic Machine Learning
Biomolecules
url https://arxiv.org/abs/2306.11715