Home › Categories › Technology› Efficient Supervised Sparse Analysis And Synthesis Operators

Efficient Supervised Sparse Analysis And Synthesis Operators.pdf

SprLitBenBroSapNIPS13.pdf

Preview of Efficient supervised sparse analysis and synthesis operators

🔗 Source: vista.cs.technion.ac.il

📊 Size: 627 KB

📄 Pages: 9 pages

⬇️ Downloads: 162

Summary

It formulates a unified approach that can be optimized for specific tasks, such as reconstruction or classification, using a bilevel optimization problem. The approach can be used to learn shift-invariant operators, which can be thought of as a way of learning sparsity-promoting convolutional operators. The paper also proposes a way of constructing feed-forward networks that can approximate the learned models at a fraction of the computational cost of exact solvers. The approach is illustrated on several experiments in music analysis and image processing applications. The main contributions of the paper are: (1) a new analysis technique for supervised learning of synthesis and analysis priors, (2) a systematic way of constructing fast fixed-complexity approximations to the solution of the proposed exact pursuit problem, and (3) a special attention to the shift-invariant particular case of the model, which leads to a principled way of constructing task-specific convolutional networks.

Download File

Description

It formulates a unified approach that can be optimized for specific tasks, such as reconstruction or classification, using a bilevel optimization problem.

Technical Information

File Format: PDF
File Size: 627 KB
Pages: 9
Language: EN
Total Downloads: 162
Last Updated: 1 month ago

Document Overview

This PDF document about Efficient supervised sparse analysis and synthesis operators provides comprehensive information and guidance. Whether you're a beginner or advanced user, this resource offers valuable insights into Efficient supervised sparse analysis and synthesis operators.