[hopos-g] CfP: Special Issue “Scientific understanding and Machine Learning in science: From traditional themes to recent developments and new vistas”
Florian Johannes Boge
Special Issue “Scientific understanding and Machine Learning in science: From traditional themes to recent developments and new vistas”, Studies in History and Philosophy of Science
We invite submissions for the virtual special issue “Scientific understanding and Machine Learning in science: From traditional themes to recent developments and new vistas”, to be published with Studies in History and Philosophy of Science.
Machine Learning (ML) systems are increasingly central to scientific inquiry, from climate modeling and drug discovery to astro- and particle physics. These developments raise fundamental questions about the nature, role, and value of understanding in science:
Does scientific understanding require interpretable models, or can it also be achieved through opaque models?
What is the role of methods from ‘explainable Artificial Intelligence’? How do these connect to traditional notions of understanding and explanation?
How do ML methods shift epistemic priorities between explanation and prediction?
Can advances brought about by ML models in science shed new light on the relation between understanding and explanation?
Are new forms of understanding with or without explanation emerging in data-intensive, ML-driven science?
How does understanding with ML models trade on traditional notions related to understanding, such as grasping, skill, or the subject, object and medium of understanding?
Historically, how did ML in science evolve from being mostly a classification device into a rich source of novel representations, knowledge and, potentially, understanding?
In this capacity, how does today’s Neural Network-driven ML compare to historically dominant forms of AI, including symbolic ones?
The VSI seeks to critically examine how ML is reshaping the epistemic landscape in science, with a particular focus on the concept of scientific understanding. It aims to provide a forum for philosophers of science and ML, as well as philosophically inclined ML practitioners, to assess whether and how ML methods are altering the conceptual and methodological foundations of scientific inquiry. By bringing these debates together in one collection, the VSI will clarify ongoing debates, identify emerging philosophical topics and frameworks, and situate ML within longer historical trajectories of scientific reasoning and inquiry.
The VSI will be open towards interdisciplinary contributions, covering both philosophical analysis and empirical case studies across a range of scientific domains where ML is currently playing a transformative role, including the physical, life and social sciences.
Manuscript submission information:
Manuscripts (up to 12,000 words) may be submitted until 20 May 2026. For any inquiries about the appropriateness of contribution topics, please contact Dr. Annika Schuster via annika....@tu-dortmund.de. For detailed information please visit https://www.sciencedirect.com/special-issue/328936/scientific-understanding-and-machine-learning-in-science-from-traditional-themes-to-recent-developments-and-new-vistas.
This email list is brought to you by HOPOS, the International Society for the History of Philosophy of Science. In order to support the list and our other activities, please consider donating to https://subfill.uchicago.edu/JournalPubs/Donation.aspx?webpub=hop
For questions about HOPOS-g, email the list master Erich Reck: erich....@gmail.com
---
You received this message because you are subscribed to the Google Groups "hop...@vt.edu" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hopos-g+u...@vt.edu.
To view this discussion visit https://groups.google.com/a/vt.edu/d/msgid/hopos-g/ca795fa7-80f3-402a-bfc9-0eb8bb969d0d%40gmail.com.