Alvarado, Ramón (2024) Challenges for Computational Reliabilism. [Preprint]

Text Challenges for Computational Reliabilism as an Epistemological Framework for ML in Science.pdf Download (339kB)

Abstract

Computational reliabilism has been recently deployed to justify our reliance and trust in computational technologies such as machine learning methods in artificial intelligence (Durán and Jongsma, 2021). Roughly, these deployments can be understood as seeking to a) respond to or circumvent the challenges related to epistemic opacity in computational methods, and in doing so, b) warrant or justify our beliefs regarding the reliability of computational processes and their results; and hence, c) to reassure us of the possibility of trust in computational methods, practices and artifacts even if these are insurmountably opaque. This chapter aims to elucidate three major challenges to computational reliabilism that have a bearing on its viability both as a general epistemological framework capable of dealing with the advent of computational methods, and as a pragmatic epistemic resolution to the justification problems related to the adoption of opaque computational methods, both of which are often cited as motivations for its adoption:

1. The challenge of warrant transmission and reliability-crediting properties
2. The challenge of the indispensability of endogenous features in artifactual reliability, and
3. The challenge of error-related opacity

---

Export/Citation:

EndNote | BibTeX | Dublin Core | ASCII/Text Citation (Chicago) | HTML Citation | OpenURL

Social Networking:

Share |

---

Item Type:	Preprint
Creators:	Creators Email ORCID Alvarado, Ramón ralvarad@uoregon.edu 0000-0002-0028-4192
Keywords:	epistemology, machine learning, reliabilism, science
Subjects:	Specific Sciences > Artificial Intelligence > AI and Ethics Specific Sciences > Computer Science Specific Sciences > Engineering Specific Sciences > Artificial Intelligence > Machine Learning General Issues > Models and Idealization General Issues > Philosophers of Science
Depositing User:	Dr. Ramón Alvarado
Date Deposited:	20 Sep 2024 11:56
Last Modified:	20 Sep 2024 11:56
Item ID:	23923
Subjects:	Specific Sciences > Artificial Intelligence > AI and Ethics Specific Sciences > Computer Science Specific Sciences > Engineering Specific Sciences > Artificial Intelligence > Machine Learning General Issues > Models and Idealization General Issues > Philosophers of Science
Date:	2024
URI:	https://philsci-archive.pitt.edu/id/eprint/23923

Monthly Views for the past 3 years

Monthly Downloads for the past 3 years

Plum Analytics

Actions (login required)

View Item