The relational bottleneck as an inductive bias for efficient abstraction
Trends in Cognitive Sciences · 2024-09 · 28 (9) · 829-843
DOI: 10.1016/j.tics.2024.04.001 ↗Kurz gefasst
Übersichtsarbeit in der Zeitschrift Trends in Cognitive Sciences, 2024-09. Verfasst von Taylor W Webb und 10 weiteren Autor:innen. Beteiligte Einrichtung: University of California, Los Angeles, CA, USA. Electronic address: taylor.w.webb@gmail.com. Hauptthemen laut MEDLINE: Concept Formation. Die Arbeit wurde bisher 24-mal zitiert (Crossref) und nennt 131 Literaturangaben. Gefördert von: National Science Foundation, Office of the Under Secretary of Defense for Research and Engineering, Office of Naval Research.
Abstract (Originalsprache)
A central challenge for cognitive science is to explain how abstract concepts are acquired from limited experience. This has often been framed in terms of a dichotomy between connectionist and symbolic cognitive models. Here, we highlight a recently emerging line of work that suggests a novel reconciliation of these approaches, by exploiting an inductive bias that we term the relational bottleneck. In that approach, neural networks are constrained via their architecture to focus on relations between perceptual inputs, rather than the attributes of individual inputs. We review a family of models that employ this approach to induce abstractions in a data-efficient manner, emphasizing their potential as candidate models for the acquisition of abstract concepts in the human mind and brain.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- University of California, Los Angeles, CA, USA. Electronic address: taylor.w.webb@gmail.com.
- Dartmouth College, Hanover, NH, USA.
- Yale University, New Haven, CT, USA.
- Princeton University, Princeton, NJ, USA.
- Brown University, Providence, RI, USA.
- University of California, Davis, CA, USA.
Förderung
- National Science Foundation
- Office of the Under Secretary of Defense for Research and Engineering
- Office of Naval Research
Zitieren
APA
Webb, T. W., Frankland, S. M., Altabaa, A., Segert, S., Krishnamurthy, K., Campbell, D., Russin, J., Giallanza, T., O'Reilly, R., Lafferty, J., & Cohen, J. D. (2024). The relational bottleneck as an inductive bias for efficient abstraction. Trends in Cognitive Sciences, 28(9), 829-843. https://doi.org/10.1016/j.tics.2024.04.001
Vancouver
Webb TW, Frankland SM, Altabaa A, Segert S, Krishnamurthy K, Campbell D, et al. The relational bottleneck as an inductive bias for efficient abstraction. Trends in Cognitive Sciences. 2024;28(9):829-843. doi:10.1016/j.tics.2024.04.001
BibTeX
@article{Webb2024,
author = {Webb, Taylor W and Frankland, Steven M and Altabaa, Awni and Segert, Simon and Krishnamurthy, Kamesh and Campbell, Declan and Russin, Jacob and Giallanza, Tyler and O'Reilly, Randall and Lafferty, John and Cohen, Jonathan D},
title = {The relational bottleneck as an inductive bias for efficient abstraction},
journal = {Trends in Cognitive Sciences},
year = {2024},
volume = {28},
number = {9},
pages = {829--843},
doi = {10.1016/j.tics.2024.04.001}
}
EUVIMED