Deep learning (DL) has become increasingly central to science, primarily due to its capacity to quickly, efficiently, and accurately predict and classify phenomena of scientific interest. This paper seeks to understand the principles that underwrite scientists’ epistemic entitlement to rely on DL in the first place and argues that these principles are philosophically novel. The question of this paper is not whether scientists can be justified in trusting in the reliability of DL. While today’s artificial intelligence exhibits characteristics common to both scientific instruments and scientific experts, this paper argues that the familiar epistemic categories that justify belief in the reliability of instruments and experts are distinct, and that belief in the reliability of DL cannot be reduced to either. Understanding what can justify belief in AI reliability represents an occasion and opportunity for exciting, new philosophy of science.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Price excludes VAT (USA)
Tax calculation will be finalised during checkout.
Philosophers have also revived interest in what we can learn about cognition from deep learning models. For instance, Buckner (2018) argues that the evaluation of the behavior of deep convolutional neural networks helps us resolve questions going back to Locke concerned with human abilities for abstraction. Others, however, have expressed skepticism about the legitimacy of looking to neural nets as plausible models of human cognition at all (Stinson, 2020).
In fact, many philosophers have argued that simulation requires special philosophical attention (Galison, 1996; Humphreys, 2004, 2009; Oreskes et al., 1994; Rohrlich, 1990; Winsberg, 2001, 2003). In general, I am sympathetic to the view that computational simulation extends the philosophical literature in genuinely fruitful ways and that consideration of simulation deepens our understanding of scientific methodology. It has, nevertheless, proved difficult to articulate precisely in what ways computational simulations give rise to specific philosophical concerns that are qualitatively distinct from those already native to the more general literature on models, experiments, or computation.
See, however, Nguyen (2020) who argues that trust is an unquestioning attitude which can be taken with respect to, among other things, ropes.
Experimental techniques are used to calibrate some physically mediated instruments. However, DLMs are not physically mediated instruments. They are mathematical functions.
Adebayo, J., Gilmer, J., Muelly, M., Goodfellow, I., Hardt, M., & Kim, B. (2018). Sanity checks for saliency maps. Advances in Neural Information Processing Systems, 31, 9505–9515.
Ashby, W. R. (1961). An introduction to cybernetics. Chapman & Hall Ltd.
Baier, A. (1986). Trust and antitrust. Ethics, 96(2), 231–260.
Baird, D. (2004). Thing knowledge: A philosophy of scientific instruments. University of California Press.
Baird, D., & Faust, T. (1990). Scientific instruments, scientific progress and the cyclotron. The British Journal for the Philosophy of Science, 41(2), 147–175.
Baker, B., Lansdell, B., Kording, K. (2021). A philosophical understanding of representation for neuroscience. arXiv preprint. arXiv:2102.06592
Baker, J. (1987). Trust and rationality. Pacific Philosophical Quarterly, 68(1), 1–13.
Birch, J., Creel, K. A., Jha, A. K., & Plutynski, A. (2022). Clinical decisions using AI must consider patient values. Nature Medicine, 28(2), 229–232.
Boge, F. J. (2021). Two dimensions of opacity and the deep learning predicament. Minds and Machines, 32(1), 43–75.
Bommasani, R., Hudson, D. A., Adeli, E., Altman, R., Arora, S., von Arx, S., Bernstein, M. S., Bohg, J., Bosselut, A., Brunskill, E., Brynjolfsson, E., Buch, S., Card, D., Castellon, R., Chatterji, N., Chen, A., Creel, K., Davis, J. Q., Demszky, D., Donahue, C., Doumbouya, M., Durmus, E., Ermon, S., Etchemendy, J., Ethayarajh, K., Fei-Fei, L., Finn, C., Gale, T., Gillespie, L., Goel, K., Goodman, N., Grossman, S., Guha, N., Hashimoto, T., Henderson, P., Hewitt, J., Ho, D. E., Hong, J., Hsu, K., Huang, J., Icard, T., Jain, S., Jurafsky, D., Kalluri, P., Karamcheti, S., Keeling, G., Khani, F., Khattab, O., Koh, P. W., Krass, M., Krishna, R., Kuditipudi, R., Kumar, A., Ladhak, F., Lee, M., Lee, T., Leskovec, J., Levent, I., Li, X. L., Li, X., Ma, T., Malik, A., Manning, C. D., Mirchandani, S., Mitchell, E., Munyikwa, Z., Nair, S., Narayan, A., Narayanan, D., Newman, B., Nie, A., Niebles, J. C., Nilforoshan, H., Nyarko, J., Ogut, G., Orr, L., Papadimitriou, I., Park, J. S., Piech, C., Portelance, E., Potts, C., Raghunathan, A., Reich, R., Ren, H., Rong, F., Roohani, Y., Ruiz, C., Ryan, J., Ré, C., Sadigh, D., Sagawa, S., Santhanam, K., Shih, A., Srinivasan, K., Tamkin, A., Taori, R., Thomas, A. W., Tramèr, F., Wang, R. E., Wang, W., Wu, B., Wu, J., Wu, Y., Xie, S. M., Yasunaga, M., You, J., Zaharia, M., Zhang, M., Zhang, T., Zhang, X., Zhang, Y., Zheng, L., Zhou, K., & Liang, P. (2021). On the opportunities and risks of foundation models. arXiv preprint. arXiv:2108.07258
Branch, B., Mirowski, P., & Mathewson, K. W. (2021). Collaborative storytelling with human actors and AI narrators. arXiv preprint. arXiv:2109.14728
Buckner, C. (2018). Empiricism without magic: Transformational abstraction in deep convolutional neural networks. Synthese, 195(12), 5339–5372.
Buckner, C. (2019). Deep learning: A philosophical introduction. Philosophy Compass, 14(10), e12625.
Charbonneau, M. (2010). Extended thing knowledge. Spontaneous Generations: A Journal for the History and Philosophy of Science, 4(1), 116–128.
Chen, Y., Lin, Z., Zhao, X., Wang, G., & Yanfeng, G. (2014). Deep learning-based classification of hyperspectral data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote sensing, 7(6), 2094–2107.
Creel, K. A. (2020). Transparency in complex computational systems. Philosophy of Science, 87(4), 568–589.
D’Amour, A., Heller, K., Moldovan, D., Adlam, B., Alipanahi, B., Beutel, A., Chen, C., Deaton, J., Eisenstein, J., Hoffman, M. D., Hormozdiari, F., Houlsby, N., Hou, S., Jerfel, G., Karthikesalingam, A., Lucic, M., Ma, Y., McLean, C., Mincu, D., Mitani, A., Montanari, A., Nado, Z., Natarajan, V., Nielson, C., Osborne, T. F., Raman, R., Ramasamy, K., Sayres, R., Schrouff, J., Seneviratne, M., Sequeira, S., Suresh, H., Veitch, V., Vladymyrov, M., Wang, X., Webster, K., Yadlowsky, S., Yun, T., Zhai, X., & Sculley, D. (2020). Underspecification presents challenges for credibility in modern machine learning. arXiv preprint. arXiv:2011.03395
Duede, E. (2022). Deep learning opacity in scientific discovery. (Forthcoming at Philosophy of Science) arXiv preprint. arXiv:2206.00520
Elgin, C. Z. (2017). True enough. MIT Press.
Engelbart, D. C. (1962). Augmenting human intellect: A conceptual framework. Menlo Park.
Falco, G., Shneiderman, B., Badger, J., Carrier, R., Dahbura, A., & Danks, D. (2021). Governing AI safety through independent audits. Nature Machine Intelligence, 3(7), 566–571.
Faulkner, P. (2007). On telling and trusting. Mind, 116(464), 875–902.
Fricker, E. (2006). Second-hand knowledge. Philosophy and Phenomenological Research, 73(3), 592–618.
Frigg, R. (2010). Fiction and scientific representation. In Beyond mimesis and convention (pp. 97–138). Springer.
Frigg, R., & Nguyen, J. (2016). The fiction view of models reloaded. The Monist, 99(3), 225–242.
Frigg, R., & Reiss, J. (2009). The philosophy of simulation: Hot new issues or same old stew? Synthese, 169(3), 593–613.
Frost-Arnold, K. (2013). Moral trust & scientific collaboration. Studies in History and Philosophy of Science Part A, 44(3), 301–310.
Galison, P. (1996). Computer simulations and the trading zone. In P. Galison & D. J. Stump (Eds.), The disunity of science: Boundaries, contexts, and power (pp. 118–157). Stanford University Press.
Galison, P. (1997). Image and logic: A material culture of microphysics. University of Chicago Press.
Gerken, M. (2015). The epistemic norms of intra-scientific testimony. Philosophy of the Social Sciences, 45(6), 568–595.
Ghorbani, A., Abid, A., & Zou, J. (2019). Interpretation of neural networks is fragile. Proceedings of the AAAI Conference on Artificial Intelligence, 33, 3681–3688.
Giere, R. N. (2010). Explaining science: A cognitive approach. University of Chicago Press.
Goldberg, S. C. (2014). Interpersonal epistemic entitlements. Philosophical Issues, 24(1), 159–183.
Goldberg, S. C. (2020). Epistemically engineered environments. Synthese, 197(7), 2783–2802.
Goldberg, S. C. (2021). What epistemologists of testimony should learn from philosophers of science. Synthese, 199(5), 12541–12559.
Goldman, A. I. (1979). What is justified belief? In Justification and knowledge (pp. 1–23). Springer.
Hacking, I. (1983). Representing and intervening: Introductory topics in the philosophy of natural science. Cambridge University Press.
Hardin, R. (1996). Trustworthiness. Ethics, 107(1), 26–42.
Hardwig, J. (1985). Epistemic dependence. The Journal of Philosophy, 82(7), 335–349.
Hardwig, J. (1991). The role of trust in knowledge. The Journal of Philosophy, 88(12), 693–708.
Harré, R. (2010). Equipment for an experiment. Spontaneous Generations: A Journal for the History and Philosophy of Science, 4(1), 30–38.
Hatherley, J. J. (2020). Limits of trust in medical AI. Journal of Medical Ethics, 46(7), 478–481.
Hieronymi, P. (2008). The reasons of trust. Australasian Journal of Philosophy, 86(2), 213–236.
Hinchman, E. S. (2005). Telling as inviting to trust. Philosophy and Phenomenological Research, 70(3), 562–587.
Holton, R. (1994). Deciding to trust, coming to believe. Australasian Journal of Philosophy, 72(1), 63–76.
Humphreys, P. (2004). Extending ourselves: Computational science, empiricism, and scientific method. Oxford University Press.
Humphreys, P. (2009). The philosophical novelty of computer simulation methods. Synthese, 169(3), 615–626.
Ioannidis, J. P. A. (2005). Why most published research findings are false. PLoS Medicine, 2(8), e124.
Jones, K. (1996). Trust as an affective attitude. Ethics, 107(1), 4–25.
Jones, K. (2012). Trustworthiness. Ethics, 123(1), 61–85.
Keren, A. (2014). Trust and belief: A preemptive reasons account. Synthese, 191(12), 2593–2615.
Khalifa, K. (2017). Understanding, explanation, and scientific knowledge. Cambridge University Press.
Lackey, J. (2010). Learning from words: Testimony as a source of knowledge. Oxford University Press.
Leavitt, M. L., & Morcos, A. (2020). Towards falsifiable interpretability research. arXiv preprint. arXiv:2010.12016
Leslie, D. (2019). Understanding artificial intelligence ethics and safety: A guide for the responsible design and implementation of AI systems in the public sector. Available at SSRN 3403301.
Lipton, Z. C. (2018). The mythos of model interpretability. Queue, 16(3), 31–57.
Meeker, K. (2004). Justification and the social nature of knowledge. Philosophy and Phenomenological Research, 69(1), 156–172.
Neyshabur, B., Tomioka, R., & Srebro, N. (2014). In search of the real inductive bias: On the role of implicit regularization in deep learning. arXiv preprint. arXiv:1412.6614
Nguyen, C. T. (2020). Trust as an unquestioning attitude. In Oxford studies in epistemology. Oxford: Oxford University Press.
Nickel, P. J. (2012). Trust and testimony. Pacific Philosophical Quarterly, 93(3), 301–316.
Nie, W., Zhang, Y., & Patel, A. (2018). A theoretical explanation for perplexing behaviors of backpropagation-based visualizations. In International conference on machine learning (pp. 3809–3818). PMLR.
Norton, S., & Suppe, F. (2001). Why atmospheric modeling is good science. In Changing the atmosphere: Expert knowledge and environmental governance (pp. 67–105).
Oreskes, N., Shrader-Frechette, K., & Belitz, K. (1994). Verification, validation, and confirmation of numerical models in the earth sciences. Science, 263(5147), 641–646.
Parker, W. S. (2008). Computer simulation through an error-statistical lens. Synthese, 163(3), 371–384.
Parker, W. S. (2008). Franklin, Holmes, and the epistemology of computer simulation. International Studies in the Philosophy of Science, 22(2), 165–183.
Parker, W. S. (2020). Model evaluation: An adequacy-for-purpose view. Philosophy of Science, 87(3), 457–477.
Rahwan, I., Cebrian, M., Obradovich, N., Bongard, J., Bonnefon, J. F., Breazeal, C., Crandall, J. W., Christakis, N. A., Couzin, I. D., Jackson, M. O., Jennings, N. R., Kamar, E., Kloumann, I. M., Larochelle, H., Lazer, D., McElreath, R., Mislove, A., Parkes, D. C., Pentland, A. S., … Wellman, M. (2019). Machine behaviour. Nature, 568(7753), 477–486.
Räz, T. (2022). Understanding deep learning with statistical relevance. Philosophy of Science, 89(1), 20–41.
Räz, T., & Beisbart, C. (2022). The importance of understanding deep learning. Erkenntnis. https://doi.org/10.1007/s10670-022-00605-y
Rohrlich, F. (1990). Computer simulation in the physical sciences. In PSA: Proceedings of the biennial meeting of the philosophy of science association (Vol. 1990, pp. 507–518). Philosophy of Science Association.
Rudin, C. (2019). Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nature Machine Intelligence, 1(5), 206–215.
Ryan, M. (2020). In AI we trust: Ethics, artificial intelligence, and reliability. Science and Engineering Ethics, 26(5), 2749–2767.
Salmon, W. C. (1971). Statistical explanation and statistical relevance (Vol. 69). University of Pittsburgh Press.
Senior, A. W., Evans, R., Jumper, J., Kirkpatrick, J., Sifre, L., Green, T., Qin, C., Žídek, A., Nelson, A. W. R., Bridgland, A., Penedones, H., Petersen, S., Simonyan, K., Crossan, S., Kohli, P., Jones, D. T., Silver, D., Kavukcuoglu, K., & Hassabi, D. (2020). Improved protein structure prediction using potentials from deep learning. Nature, 577(7792), 706–710.
Shapin, S., & Schaffer, S. (2011). Leviathan and the air-pump. Princeton University Press.
Sines, G., & Sakellarakis, Y. A. (1987). Lenses in antiquity. American Journal of Archaeology, 91, 191–196.
Smith, P. J., & Hoffman, R. R. (2017). Cognitive systems engineering: The future for a changing world. Crc Press.
Sourati, J., & Evans, J. (2021). Accelerating science with human versus alien artificial intelligences. arXiv preprint. arXiv:2104.05188
Stevens, R., Taylor, V., Nichols, J., Maccabe, A. B., Yelick, K., & Brown, D. (2020). AI for science. Technical report, Argonne National Lab.(ANL), Argonne.
Stinson, C. (2020). From implausible artificial neurons to idealized cognitive models: Rebooting philosophy of artificial intelligence. Philosophy of Science, 87(4), 590–611.
Sullivan, E. (2019). Understanding from machine learning models. British Journal for the Philosophy of Science. https://doi.org/10.1093/bjps/axz035
Wang, S., Kai, F., Luo, N., Cao, Y., Wu, F., Zhang, C., Heller, K. A, & You, L. (2019). Massive computational acceleration by using neural networks to emulate mechanism-based biological models. bioRxiv (p. 559559).
Weisberg, M. (2012). Simulation and similarity: Using models to understand the world. Oxford University Press.
Wilholt, T. (2020). Epistemic trust in science. The British Journal for the Philosophy of Science. https://doi.org/10.1093/bjps/axs007
Winsberg, E. (2001). Simulations, models, and theories: Complex physical systems and their representations. Philosophy of Science, 68(S3), S442–S454.
Winsberg, E. (2003). Simulated experiments: Methodology for a virtual world. Philosophy of Science, 70(1), 105–125.
Winsberg, E. (2010). Science in the age of computer simulation. University of Chicago Press.
Zerilli, J. (2022). Explaining machine learning decisions. Philosophy of Science, 89(1), 1–19.
Zhang, C., Bengio, S., Hardt, M., Recht, B., & Vinyals, O. (2021). Understanding deep learning (still) requires rethinking generalization. Communications of the ACM, 64(3), 107–115.
Zik, Y., & Hon, G. (2017). History of science and science combined: Solving a historical problem in optics—The case of Galileo and his telescope. Archive for History of Exact Sciences, 71(4), 337–344.
This manuscript benefited greatly from conversations with Kevin Davey, Tyler Millhouse, Jennifer Nagel, Wendy Parker, Tom Pashby, Anubav Vasudevan, Bill Wimsatt, participants of the Theoretical Philosophy Workshop at the University of Chicago, and the insightful feedback of two anonymous referees.
This work was supported by the US National Science Foundation #2022023 NRT-HDR: AI-enabled Molecular Engineering of Materials and Systems (AIMEMS) for Sustainability.
Conflict of interest
The authors declare that they have no conflicts of interest, including affiliation with or involvement in an organization or entity with a financial or non-financial interest in the subject matter or materials discussed in this manuscript.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This paper is forthcoming in a special issue of Synthese titled “Philosophy of Science in Light of Artificial Intelligence”.
T.C. : Philosophy of Science in Light of Artificial Intelligence.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Duede, E. Instruments, agents, and artificial intelligence: novel epistemic categories of reliability. Synthese 200, 491 (2022). https://doi.org/10.1007/s11229-022-03975-6
- Deep learning
- Scientific knowledge
- Trust and Justification