The DOI matches its Crossref record; 95 of the 98 references it cites were found in a registry or an index.
Empiricism without magic: transformational abstraction in deep convolutional neural networks
Open accessAbstract
In artificial intelligence, recent research has demonstrated the remarkable potential of Deep Convolutional Neural Networks (DCNNs), which seem to exceed state-of-the-art performance in new domains weekly, especially on the sorts of very difficult perceptual discrimination tasks that skeptics thought would remain beyond the reach of artificial intelligence. However, it has proven difficult to explain why DCNNs perform so well. In philosophy of mind, empiricists have long suggested that complex cognition is based on information derived from sensory experience, often appealing to a faculty of abstraction. Rationalists have frequently complained, however, that empiricists never adequately explained how this faculty of abstraction actually works. In this paper, I tie these two questions together, to the mutual benefit of both disciplines. I argue that the architectural features that distinguish DCNNs from earlier neural networks allow them to implement a form of hierarchical processing that I call “transformational abstraction”. Transformational abstraction iteratively converts sensory-based representations of category exemplars into new formats that are increasingly tolerant to “nuisance variation” in input. Reflecting upon the way that DCNNs leverage a combination of linear and non-linear processing to efficiently accomplish this feat allows us to understand how the brain is capable of bi-directional travel between exemplars and abstractions, addressing longstanding problems in empiricist philosophy of mind. I end by considering the prospects for future research on DCNNs, arguing that rather than simply implementing 80s connectionism with more brute-force computation, transformational abstraction counts as a qualitatively distinct form of processing ripe with philosophical and psychological significance, because it is significantly better suited to depict the generic mechanism responsible for this important kind of psychological processing in the brain.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“As Buckner [5] argues, contemporary AI systems operate in terms of empiricism by abstracting complex patterns from data bases, in which performance improvements are largely driven not by conceptual breakthroughs in representation or reasoning, but by the scale and character of empirical data.”
From 4E cognition and the coevolution of human–AI interaction (2025) · cites this paper as [5]
Citations
Cites
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 73,472
Human-level control through deep reinforcement learning
Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A. A., Veness, J., Bellemare, M. G., Graves, A., Riedmiller, M., Fidjeland, A. K., Ostrovski, G., et al. (2015) · Nature · doi:10.1038/nature14236 · Cited by 30,357
Reducing the dimensionality of data with neural networks
Hinton, G. E., & Salakhutdinov, R. R. (2006) · Science · doi:10.1126/science.1127647 · Cited by 20,928
Deep learning in neural networks: an overview
Schmidhuber, J. (2015) · Neural Networks · doi:10.1016/j.neunet.2014.09.003 · Cited by 18,570 · cited 4×
“…was the way it com- 7 In the interests of space, we move quickly over the history here; for more background and discussion, see (Buckner and Garson 2018; Schmidhuber 2015). bined two different types of operation—linear convolutional filtering and nonlinear downsampling—in a single network.”
BackgroundMethodInfluentialMastering the game of Go with deep neural networks and tree search
Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., van den Driessche, G., Schrittwieser, J., Antonoglou, I., Panneershelvam, V., Lanctot, M., et al. (2016) · Nature · doi:10.1038/nature16961 · Cited by 15,612
Approximation by superpositions of a sigmoidal function
Cybenko, G. (1989) · Mathematics of Control, Signals, and Systems · doi:10.1007/bf02551274 · Cited by 13,883 · cited 1×
“…of mathematical abstraction, all neural networks perform pattern recognition or dimensionality reduction, and the universal approximation theory shows that in principle a three-layer perceptron with an unlimited number of nodes can approximate any continuous function (Cybenko 1989; Hornik 1991).”
BackgroundReceptive fields, binocular interaction and functional architecture in the cat’s visual cortex
Hubel, D. H., & Wiesel, T. N. (1962) · The Journal of Physiology · doi:10.1113/jphysiol.1962.sp006837 · Cited by 13,680 · cited 1×
“This traditionwas inspired by an anatomical discovery in cat visual cortex by Hubel and Wiesel (1962).”
MethodBackpropagation Applied to Handwritten Zip Code Recognition
LeCun, Y., Boser, B., Denker, J. S., Henderson, D., Howard, R. E., Hubbard, W., & Jackel, L. D. (1989) · Neural Computation · doi:10.1162/neco.1989.1.4.541 · Cited by 11,891 · cited 1×
“Despite some early successes (LeCun et al. 1989), when backpropagation was applied to such deep networks, learning tended to settle into poor localminima or fail to stabilize at all (DeMers andCottrell 1993).”
Background(2016) · Cited by 11,679
Research works printed but not linked: 7
- Blundell, C., Uria, B., Pritzel, A., Li, Y., Ruderman, A., Leibo, J. Z., et al. (2016). Model-free episodic control. arXiv Preprint arXiv:1606.04460 . · arXiv
- Elsayed, G. F., Shankar, S., Cheung, B., Papernot, N., Kurakin, A., Goodfellow, I., & Sohl-Dickstein, J. (2018). Adversarial examples that fool both human and computer vision. arXiv Preprint arXiv:1802.08195 . · arXiv
- Gray, H. (1918). Anatomy of the human body, rev. and re-edited by Warren H. Lewis. Philadelphia: Lea & Febiger.
- Hochreiter, S. (1991). Untersuchungen zu dynamischen neuronalen Netzen. Diploma thesis for Technische Universität München, München.
- Hume, D. (1739). A treatise on human nature. Oxford: Oxford University Press.
- Marcus, G. (2018). Deep learning: A critical appraisal. arXiv:1801.00631 [cs, Stat]. · arXiv
- Quine, W. V. (1971). Epistemology naturalized. Akten Des XIV. Internationalen Kongresses Für Philosophie, 6, 87–103.
Cited by
Solving the Black Box Problem: A Normative Framework for Explainable Artificial Intelligence
Zednik, C. (2021) · Philosophy & Technology · doi:10.1007/s13347-019-00382-7 · Cited by 419
“Rather, the system is more likely to deploy a taxonomy of abstract categories in which the inputs are combined nonlinearly (see also Buckner 2018).”
BackgroundInfluentialUnderstanding from Machine Learning Models
Sullivan, E. (2022) · The British Journal for the Philosophy of Science · doi:10.1093/bjps/axz035 · Cited by 221
“See Buckner ([2018]) for a philosophical argument that DNNs (CNNs in particular) capture processes of the brain.”
BackgroundDeep learning: A philosophical introduction
Buckner, C. J. (2019) · Philosophy Compass · doi:10.1111/phc3.12625 · Cited by 194
“…knowledge, a faculty that has frequently been invoked by empiricists to account for learners' transition from basic sensory associations to the abstract representations deployed in higher cognition but has until recently not been adequately explained (c.f. Buckner, 2018; Laurence & Margolis, 2012).”
BackgroundSelf-citationTwo Dimensions of Opacity and the Deep Learning Predicament
Boge, F. J. (2021) · Minds and Machines · doi:10.1007/s11023-021-09569-4 · Cited by 171
“Note that the learning algorithm involved in deriving this model may itself count as yet another model: Buckner (2018) points to the possibility of understanding concept abstraction on the basis of deep convolutional nets, without drawing too close a parallel to either brain processes or most…”
Neurocognitive Mechanisms: Explaining Biological Cognition
Piccinini, G. (2020) · doi:10.1093/oso/9780198866282.001.0001 · Cited by 162
Metadata
- Authors
- Buckner, C.
- Affiliations
- University of Houston (US)
- Corresponding author
- Buckner, Cameron
- ORCID
- Buckner, Cameron
- Venue
- Synthese
- Published
- 2018-09-24
- Volume
- 195
- Issue
- 12
- Pages
- 5339-5372
- ISSN
- 0039-7857 (ISSN-L)
- 1573-0964
- OpenAlex
- W2889853714
- Cited by
- 149
- References
- 98
- Citations by year
- 2026: 5 · 2025: 13 · 2024: 17 · 2023: 28 · 2022: 17 · 2021: 26 · 2020: 33 · 2019: 9 · 2018: 1
- FWCI
- 22.60
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- www.springer.com/tdm (tdm)
- Locations
- Synthese (publishedVersion, cc-by)
- PhilSci-Archive (University of Pittsburgh) (acceptedVersion)
- PhilPapers (PhilPapers Foundation) (submittedVersion)
- Topics
- Social Sciences › Arts and Humanities › Philosophy › Philosophy and Theoretical Science
- Social Sciences › Arts and Humanities › History and Philosophy of Science › Cybernetics and Technology in Society
- Life Sciences › Neuroscience › Cognitive Neuroscience › Embodied and Extended Cognition
- Keywords
- deep convolutional neural networks, empiricism, rationalism, hierarchical processing, perceptual discrimination, connectionism, philosophy of mind, neural network architecture
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Empiricism without magic: transformational abstraction in deep convolutional neural networks. https://app.cochat.ai/papers/10.1007/s11229-018-01949-1
- The paper
- Buckner, C. (2018). Empiricism without magic: transformational abstraction in deep convolutional neural networks. Synthese. https://doi.org/10.1007/s11229-018-01949-1
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.