What can 1.8 billion regressions tell us about the pressures shaping high-level visual representation in brains and machines?

C Conwell, JS Prince, KN Kay, GA Alvarez, T Konkle - BioRxiv, 2022 - biorxiv.org
The rapid development and open-source release of highly performant computer vision
models offers new potential for examining how different inductive biases impact …

A large-scale examination of inductive biases shaping high-level visual representation in brains and machines

C Conwell, JS Prince, KN Kay, GA Alvarez… - Nature …, 2024 - nature.com
The rapid release of high-performing computer vision models offers new potential to study
the impact of different inductive biases on the emergent brain alignment of learned …

Layerwise complexity-matched learning yields an improved model of cortical area V2

N Parthasarathy, OJ Hénaff, EP Simoncelli - ArXiv, 2024 - pmc.ncbi.nlm.nih.gov
Human ability to recognize complex visual patterns arises through transformations
performed by successive areas in the ventral visual cortex. Deep neural networks trained …

FroSSL: Frobenius Norm Minimization for Efficient Multiview Self-supervised Learning

O Skean, A Dhakal, N Jacobs… - European Conference on …, 2025 - Springer
Self-supervised learning (SSL) is a popular paradigm for representation learning. Recent
multiview methods can be classified as sample-contrastive, dimension-contrastive, or …

Towards an Improved Understanding and Utilization of Maximum Manifold Capacity Representations

R Schaeffer, V Lecomte, DB Pai, A Carranza… - arXiv preprint arXiv …, 2024 - arxiv.org
Maximum Manifold Capacity Representations (MMCR) is a recent multi-view self-supervised
learning (MVSSL) method that matches or surpasses other leading MVSSL methods. MMCR …

An information-theoretic understanding of maximum manifold capacity representations

V Lecomte, R Schaeffer, B Isik, M Khona… - … 2023 Workshop on …, 2023 - openreview.net
Maximum Manifold Capacity Representations (MMCR) is a recent multi-view self-supervised
learning (MVSSL) method that matches or surpasses other leading MVSSL methods. MMCR …

An information-theoretic understanding of maximum manifold capacity representations

B Isik, V Lecomte, R Schaeffer, Y LeCun… - UniReps: the First …, 2023 - openreview.net
Maximum Manifold Capacity Representations (MMCR) is a recent multi-view self-supervised
learning (MVSSL) method that matches or surpasses other leading MVSSL methods. MMCR …

Non-agriculturalization Detection Based on Vector Polygons and Contrastive Learning with High Resolution Remote Sensing Images

H Zhang, W Liu, C Zhu, H Niu, P Yin… - IEEE Journal of …, 2024 - ieeexplore.ieee.org
The conversion of agricultural lands, termed “nonagriculturalization,” poses profound threats
to food security and ecological stability. Remote sensing image change detection offers an …

Learning predictable and robust neural representations by straightening image sequences

X Niu, C Savin, EP Simoncelli - arXiv preprint arXiv:2411.01777, 2024 - arxiv.org
Prediction is a fundamental capability of all living organisms, and has been proposed as an
objective for learning sensory representations. Recent work demonstrates that in primate …

Improving Pre-Trained Self-Supervised Embeddings Through Effective Entropy Maximization

D Chakraborty, Y LeCun, TGJ Rudner… - arXiv preprint arXiv …, 2024 - arxiv.org
A number of different architectures and loss functions have been applied to the problem of
self-supervised learning (SSL), with the goal of developing embeddings that provide the …