Year: 2018
Type: conference-paper
Abstract: Inner product-based convolution has been a central component of convolutional neural networks (CNNs) and the key to learning visual representations. Inspired by the observation that CNN-learned features are naturally decoupled with the norm of features corresponding to the intra-class variation and the angle corresponding to the semantic difference, we propose a generic decoupled learning framework which models the intra-class variation and semantic difference independently. Specifically, we first reparametrize the inner product to a decoupled form and then generalize it to the decoupled convolution operator which serves as the building block of our decoupled networks. We present several effective instances of the decoupled convolution operator. Each decoupled operator is well motivated and has an intuitive geometric interpretation. Based on these decoupled operators, we further propose to directly learn the operator from data. Extensive experiments show that such decoupled reparameterization renders significant performance gain with easier convergence and stronger robustness.
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: G1843612087, III: Small: Collaborative Research: Efficient, Nonparametric and Local-Minimum-Free Latent Variable Models: With Application to Large-Scale Computer Vision and Genomics, Comp Cog: Collaborative Research on the Development of Visual Object Recognition, EAGER: Asynchronous Event Models for State-Topology Co-Evolution of Temporal Networks, SaTC: CORE: Medium: Understanding and Fortifying Machine Learning Based Security Analytics, CAREER: New Representations of Probability Distributions to Improve Machine Learning --- A Unified Kernel Embedding Framework for Distributions
OpenAlex ID: W4232886157
