SOPhiA 2021

Salzburgiense Concilium Omnibus Philosophis Analyticis

SOPhiA ToolsDE-pageEN-page

Programme - Talk

Epistemic risks and computer simulation: a case study from particle physics
(Philosophy of Science, )

In philosophy of sciences, the issue of epistemic risk is usually addressed in terms of inductive risk, focusing on the process of decision-making to accept or reject hypotheses based on empirical evidence. This topic is widely discussed in the literature on the Argument from Inductive Risk (AIR) (Steel, 2010) and mainly concerns with the role of value-laden judgements in weighing evidence to prevent from social and ethical harm.

In many sciences today, however, empirical reasoning is highly inferential as experiments rely on complex instrumented disposals. This means that there is a long process before confronting evidence to hypothesis. This process often involves an increasing use of computer simulations, may it be in life science or particle physics where computer simulations are, for example, centrally involved in the design of particle detectors and data generation. The crucial role of these computer-based practices, which are in this context precondition for empirical reasoning, call for further philosophical insight regarding risks.

In this paper, we zoom in on particle physics and aim to expand the framework of epistemic risks to particularly address the issue of computer simulation-related risks. Based on a case study from ATLAS experiment in top-quark physics we argue that there are relevant epistemic risks besides inductive ones that go beyond social and ethical impacts. The subsumption of risks under inductive ones is insufficient to address the variety of risk arising in the course of scientific inquiry as well as to address the collaborative feature of producing scientific knowledge (Biddle & Kukla, 2017). After analyzing contingent choices made in the experimental process, we propose to frame epistemic risk as the risk to not fulfil one's epistemic aim, distinguishing between local (e.g., prediction) and global (e.g., discovery) aims. Our contribution can be understood as an attempt to locate uncertainty and risks and explicate relationships at stake.

CITED REFERENCES

Biddle, J. B., & Kukla, R. (2017). The Geography of Epistemic Risk. In Exploring Inductive Risk: Case Studies of Values in Science (Vol. 1). Oxford University Press. https://doi.org/10.1093/acprof:oso/9780190467715.003.0011

Steel, D. (2010). Epistemic Values and the Argument from Inductive Risk*. Philosophy of Science, 77(1), 14 34. https://doi.org/10.1086/650206


Chair: Andelija Milic
Time: 14:00-14:30, 09 September 2021 (Thursday)
Location: SR 1.006
Remark: (Online Talk)

Marianne van Panhuys 
(Karlsruhe Institute for Technology, Germany)

Currently located at Karlsruhe Institute of Technology in the research group PhilETAS, I am a doctoral student working on the subproject ''Impact of Computer Simulation and Machine Learning on the epistemic status of LHC data'' of the interdisciplinary research unit "Epistemology of the Large Hadron Colider". My research focuses on epistemic risks induced by computer simulation in the ATLAS experiment.

I have completed my Master degree in "Logic, History and Philosophy of Science and Technology" at the University of Lyon in France. I've been working on Karl Popper's cosmological thesis of propensity and on predictive technology in contemporary science. During my training I was lucky to do an internship at the Institute of Physics of Lyon (IP2I) with an emphasis on CMS (Compact Muon Solenoid) experiment of the LHC.



Testability and Meaning deco