What question did this study set out to answer?

The aim is to understand how data representations can be both invariant to transformations and selective, meaning similar inputs have corresponding representations.

May 9, 2016Open Access

On invariance and selectivity in representation learning

Key Points

The aim is to understand how data representations can be both invariant to transformations and selective, meaning similar inputs have corresponding representations.
Mathematical analysis of invariance and selectivity in representation learning
Connection to i-theory and feedforward processing of sensory data
Review of theoretical claims regarding deep architectures in learning.
Demonstrated that effective representations can be created that meet invariance and selectivity criteria.
Strengthened theoretical claims from previous studies on sensory representation learning.
Provided insights relevant to deep learning architectures and sensory cortex functionalities.

Abstract

We study the problem of learning from data representations that are invariant to transformations, and at the same time selective, in the sense that two points have the same representation if one is the transformation of the other. The mathematical results here sharpen some of the key claims of i-theory—a recent theory of feedforward processing in sensory cortex (Anselmi et al., 2013, Theor. Comput. Sci. and arXiv:1311.4158; Anselmi et al., 2013, Magic materials: a theory of deep hierarchical architectures for learning sensory representations. CBCL Paper; Anselmi & Poggio, 2010, Representation learning in sensory cortex: a theory. CBMM Memo No. 26).

Read Full Paperexternally

Bookmark

View Full Paper

Cite This Study

Anselmi et al. (Mon,) studied this question.

synapsesocial.com/papers/6a0f2d7cf27f69a1d3426311 https://doi.org/https://doi.org/10.1093/imaiai/iaw009

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

Bookmark

View Full Paper