Teasing Apart Architecture and Initial Weights as Sources of Inductive Bias in Neural Networks

Gianluca M. Bencomo, Max Gupta, Ioana Marinescu, R. Thomas McCoy, Tom Griffiths 0001. Teasing Apart Architecture and Initial Weights as Sources of Inductive Bias in Neural Networks. In David Barner, Neil Bramley, Azzurra Ruggeri, Caren M. Walker, editors, Proceedings of the 47th Annual Meeting of the Cognitive Science Society, CogSci 2025, San Francisco, CA, USA, July 30 - August 2, 2025. cognitivesciencesociety.org, 2025. [doi]

Abstract

Abstract is missing.