On the Level Sets and Invariance of Neural Tuning Landscapes
Preprint
Abstract
Visual representations can be defined as the activations of neuronal populations in response to images. The activation of a neuron as a function over all image space has been described as a "tuning landscape". As a function over a high-dimensional space, what is the structure of this landscape? In this study, we characterize tuning landscapes through the lens of level sets and Morse theory. A recent study measured the in vivo two-dimensional tuning maps of neurons in different brain regions. Here, we developed a statistically reliable signature for these maps based on the change of topology in level sets. We found this topological signature changed progressively throughout the cortical hierarchy, with similar trends found for units in convolutional neural networks (CNNs). Further, we analyzed the geometry of level sets on the tuning landscapes of CNN units. We advanced the hypothesis that higher-order units can be locally regarded as isotropic radial basis functions, but not globally. This shows the power of level sets as a conceptual tool to understand neuronal activations over image space.Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- High-performance evolutionary algorithms for online neuron control2022
- Neural tuning and representational geometryNature Reviews Neuroscience, 2021
- Separability and geometry of object manifolds in deep neural networksNature Communications, 2020
- Evolving Images for Visual Neurons Using a Deep Generative Network Reveals Coding Principles and Neuronal PreferencesCell, 2019
- UMAP: Uniform Manifold Approximation and ProjectionJournal of Open Source Software, 2018
- Classification and Geometry of General Perceptual ManifoldsPhysical Review X, 2018
- A computationally efficient limited memory CMA-ES for large scale optimization2014
- Spatiotemporal Elements of Macaque V1 Receptive FieldsNeuron, 2005
- Level Set Methods and Dynamic Implicit Surfaces2003
- Energy Integration Describes Sound-Intensity Coding in an Insect Auditory SystemJournal of Neuroscience, 2002