Simplicity Bias in Overparameterized Machine Learning

Research output: Contribution to journalConference articlepeer-review


A thorough theoretical understanding of the surprising generalization ability of deep networks (and other overparameterized models) is still lacking. Here we demonstrate that simplicity bias is a major phenomenon to be reckoned with in overparameterized machine learning. In addition to explaining the outcome of simplicity bias, we also study its source: following concrete rigorous examples, we argue that (i) simplicity bias can explain generalization in overparameterized learning models such as neural networks; (ii) simplicity bias and excellent generalization are optimizer-independent, as our examples show, and although the optimizer affects training, it is not the driving force behind simplicity bias; (iii) simplicity bias in pre-training models, and subsequent posteriors, is universal and stems from the subtle fact that uniformly-at-random constructed priors are not uniformly-at-random sampled; and (iv) in neural network models, the biasing mechanism in wide (and shallow) networks is different from the biasing mechanism in deep (and narrow) networks.

Original languageEnglish
Pages (from-to)11052-11060
Number of pages9
JournalProceedings of the AAAI Conference on Artificial Intelligence
Issue number10
StatePublished - 25 Mar 2024
Event38th AAAI Conference on Artificial Intelligence, AAAI 2024 - Vancouver, Canada
Duration: 20 Feb 202427 Feb 2024

ASJC Scopus subject areas

  • Artificial Intelligence


Dive into the research topics of 'Simplicity Bias in Overparameterized Machine Learning'. Together they form a unique fingerprint.

Cite this