Diverse explanations from data-driven and domain-driven perspectives in the physical sciences
Machine learning methods have been remarkably successful in material science, providing novel scientific insights, guiding future laboratory experiments, and accelerating materials discovery. Despite the promising performance of these models, understanding the decisions they make is also essential t...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2025-01-01
|
| Series: | Machine Learning: Science and Technology |
| Subjects: | |
| Online Access: | https://doi.org/10.1088/2632-2153/ad9137 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850025444166860800 |
|---|---|
| author | Sichao Li Xin Wang Amanda Barnard |
| author_facet | Sichao Li Xin Wang Amanda Barnard |
| author_sort | Sichao Li |
| collection | DOAJ |
| description | Machine learning methods have been remarkably successful in material science, providing novel scientific insights, guiding future laboratory experiments, and accelerating materials discovery. Despite the promising performance of these models, understanding the decisions they make is also essential to ensure the scientific value of their outcomes. However, there is a recent and ongoing debate about the diversity of explanations, which potentially leads to scientific inconsistency. This Perspective explores the sources and implications of these diverse explanations in ML applications for physical sciences. Through three case studies in materials science and molecular property prediction, we examine how different models, explanation methods, levels of feature attribution, and stakeholder needs can result in varying interpretations of ML outputs. Our analysis underscores the importance of considering multiple perspectives when interpreting ML models in scientific contexts and highlights the critical need for scientists to maintain control over the interpretation process, balancing data-driven insights with domain expertise to meet specific scientific needs. By fostering a comprehensive understanding of these inconsistencies, we aim to contribute to the responsible integration of eXplainable artificial intelligence into physical sciences and improve the trustworthiness of ML applications in scientific discovery. |
| format | Article |
| id | doaj-art-882e28bd2f1849878c827c89d0c76da2 |
| institution | DOAJ |
| issn | 2632-2153 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | Machine Learning: Science and Technology |
| spelling | doaj-art-882e28bd2f1849878c827c89d0c76da22025-08-20T03:00:50ZengIOP PublishingMachine Learning: Science and Technology2632-21532025-01-016101300210.1088/2632-2153/ad9137Diverse explanations from data-driven and domain-driven perspectives in the physical sciencesSichao Li0https://orcid.org/0000-0002-0097-6754Xin Wang1Amanda Barnard2https://orcid.org/0000-0002-4784-2382School of Computing, Australian National University , Canberra, AustraliaSchool of Computing, Australian National University , Canberra, AustraliaSchool of Computing, Australian National University , Canberra, AustraliaMachine learning methods have been remarkably successful in material science, providing novel scientific insights, guiding future laboratory experiments, and accelerating materials discovery. Despite the promising performance of these models, understanding the decisions they make is also essential to ensure the scientific value of their outcomes. However, there is a recent and ongoing debate about the diversity of explanations, which potentially leads to scientific inconsistency. This Perspective explores the sources and implications of these diverse explanations in ML applications for physical sciences. Through three case studies in materials science and molecular property prediction, we examine how different models, explanation methods, levels of feature attribution, and stakeholder needs can result in varying interpretations of ML outputs. Our analysis underscores the importance of considering multiple perspectives when interpreting ML models in scientific contexts and highlights the critical need for scientists to maintain control over the interpretation process, balancing data-driven insights with domain expertise to meet specific scientific needs. By fostering a comprehensive understanding of these inconsistencies, we aim to contribute to the responsible integration of eXplainable artificial intelligence into physical sciences and improve the trustworthiness of ML applications in scientific discovery.https://doi.org/10.1088/2632-2153/ad9137XAIphysical scienceexplanations |
| spellingShingle | Sichao Li Xin Wang Amanda Barnard Diverse explanations from data-driven and domain-driven perspectives in the physical sciences Machine Learning: Science and Technology XAI physical science explanations |
| title | Diverse explanations from data-driven and domain-driven perspectives in the physical sciences |
| title_full | Diverse explanations from data-driven and domain-driven perspectives in the physical sciences |
| title_fullStr | Diverse explanations from data-driven and domain-driven perspectives in the physical sciences |
| title_full_unstemmed | Diverse explanations from data-driven and domain-driven perspectives in the physical sciences |
| title_short | Diverse explanations from data-driven and domain-driven perspectives in the physical sciences |
| title_sort | diverse explanations from data driven and domain driven perspectives in the physical sciences |
| topic | XAI physical science explanations |
| url | https://doi.org/10.1088/2632-2153/ad9137 |
| work_keys_str_mv | AT sichaoli diverseexplanationsfromdatadrivenanddomaindrivenperspectivesinthephysicalsciences AT xinwang diverseexplanationsfromdatadrivenanddomaindrivenperspectivesinthephysicalsciences AT amandabarnard diverseexplanationsfromdatadrivenanddomaindrivenperspectivesinthephysicalsciences |