Classification and Geometry of General Perceptual Manifolds

2017/10/31 by SueYeon Chung, Daniel D. Lee, Haim Sompolinsky · 1 voice · 136 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · Engineering · Mathematics · Neuroscience · Physics and Astronomy · Psychology · #Computer science #Engineering #Face and Expression Recognition #Geometry #Manifold (fluid mechanics) #Mathematics #Neural Networks and Applications #Perception #Psychology #Riemannian geometry #Visual perception and processing mechanisms #cond-mat.dis-nn #cond-mat.stat-mech #cs.NE #q-bio.NC #stat.ML

paper · pdf · doi:10.1103/physrevx.8.031003

published in Physical Review X 8(3) (American Physical Society) · 24 pages, 12 figures, Supplementary Materials

arxiv created 2018/06/24 · openalex publication_date 2018/07/05 · arxiv updated 2018/07/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06

Abstract

Perceptual manifolds arise when a neural population responds to an ensemble of sensory signals associated with different physical features (e.g., orientation, pose, scale, location, and intensity) of the same perceptual object. Object recognition and discrimination require classifying the manifolds in a manner that is insensitive to variability within a manifold. How neuronal systems give rise to invariant object classification and recognition is a fundamental problem in brain theory as well as in machine learning.

Cited by

• The cortical critical power law balances energy and information in an optimal fashion
• Conditioned Direct Feedback Alignment via Activity and Error Geometry
• Why the Brain Consolidates: Predictive Forgetting for Optimal Generalisation
• From sensory to perceptual manifolds: The twist of neural geometry
• The Geometry of Prompting: Unveiling Distinct Mechanisms of Task Adaptation in Language Models
• Intrinsic dimension of data representations in deep neural networks
• Neural tuning and representational geometry
• Machine learning and the physical sciences
• Computational role of structure in neural activity and connectivity
• Dimensionality compression and expansion in Deep Neural Networks
• Readout Representation: Redefining Neural Codes by Input Recovery
• Estimating Dimensionality of Neural Representations from Finite Samples
• Interpreting Encoding and Decoding Models
• Microscopic and collective signatures of feature learning in neural networks
• Dynamical Mechanism of Sampling-based Stochastic Inference under Probabilistic Population Codes
• Understanding the Logit Distributions of Adversarially-Trained Deep Neural Networks
• Contrastive Self-Supervised Learning As Neural Manifold Packing
• Interpreting encoding and decoding models
• Statistical mechanics of extensive-width Bayesian neural networks near interpolation
• Biologically inspired architectures for sample-efficient deep reinforcement learning
• On the geometry of generalization and memorization in deep neural networks
• What's Inside Your Diffusion Model? A Score-Based Riemannian Metric to Explore the Data Manifold
• Emergence of Separable Manifolds in Deep Language Representations
• Convolutional Neural Networks as a Model of the Visual System: Past, Present, and Future
• Do Sparse Autoencoders Capture Concept Manifolds?
• Neural population geometry: An approach for understanding biological and artificial neural networks
• Structuring Sparsity: Block-Sparse Featurizers Capture Visual Concept Manifolds
• Large language models reorganize representational geometry during in-context learning
• Decoding Alignment without Encoding Alignment: A critique of similarity analysis in neuroscience
• Summary statistics of learning link changing neural representations to behavior

Discussions

• Enjoyed speaking at Frontiers in NeuroAI Symposium @harvard.edu's @kempnerinstitute.bsky.social Key papers from the talk: *Manifold capacity theory: - original: doi.org/10.1103/Phys... - correlated [bsky, 26 points, 1 comments]

Related

• Statistical Mechanics of Neural Processing of Object Manifolds
• From sensory to perceptual manifolds: The twist of neural geometry
• Geometric Origins of Bias in Deep Neural Networks: A Human Visual System Perspective
• Quantifying Extrinsic Curvature in Neural Manifolds
• Investigating Image Manifolds of 3D Objects: Learning, Shape Analysis, and Comparisons