Campus Units

Mechanical Engineering, Electrical and Computer Engineering, Plant Sciences Institute

Document Type

Conference Proceeding

Conference

NeurIPS Thirty-fourth Annual Conference on Neural Information Processing Systems (NeuroIPS 2020)

Publication Version

Published Version

Publication Date

2020

Conference Title

NeurIPS Thirty-fourth Annual Conference on Neural Information Processing Systems (NeurIPS 2020)

Conference Date

December 6-12, 2020

Abstract

We introduce a new, principled approach to extend gradient-based optimization to piecewise smooth models, such as k-histograms, splines, and segmentation maps. We derive an accurate form of the weak Jacobian of such functions and show that it exhibits a block-sparse structure that can be computed implicitly and efficiently. We show that using the redesigned Jacobian leads to improved performance in applications such as denoising with piecewise polynomial regression models, datafree generative model training, and image segmentation.

Comments

This proceeding is published as Cho, Minsu, Ameya Joshi, Xian Yeow Lee, Aditya Balu, Adarsh Krishnamurthy, Baskar Ganapathysubramanian, Soumik Sarkar, and Chinmay Hegde. "Differentiable Programming for Piecewise Polynomial Functions." NeurIPS Thirty-fourth Annual Conference on Neural Information Processing Systems. Learning Meets Combinatorial Algorithms (LMCA): Workshop at NeurIPS 2020. December 6-12, 2020. Posted with permission.

Copyright Owner

The Author(s)

Language

en

File Format

application/pdf

Share

 
COinS