Sparsifying Parametric Models with L0 Regularization

Nicolò Botteghi, Urban Fasel
9/5/2024
cs.LG

Official code: nicob15/Sparsifying-Parametric-Models-with-L0 (5 stars, MIT, last commit 2024-10-23).

Abstract

This document contains an educational introduction to the problem of sparsifying parametric models with L0 regularization. We utilize this approach together with dictionary learning to learn sparse polynomial policies for deep reinforcement learning to control parametric partial differential equations. The code and a tutorial are provided here: https://github.com/nicob15/Sparsifying-Parametric-Models-with-L0.

View on arXivView PDF

Code Implementations(1)

nicob15/Sparsifying-Parametric-Models-with-L0Official100%
MIT
51Python, Jupyter NotebookMay 17, 20241 years agoMIT

Similar Papers

Parametric PDE Control with Deep Reinforcement Learning and Differentiable L0-Sparse Polynomial Policies

86% match2 repos52024

Learning sparsity-promoting regularizers for linear inverse problems

83% match2024

Optimal Policy Sparsification and Low Rank Decomposition for Deep Reinforcement Learning

82% match2024

Sparse Deep Learning Models with the $\ell_1$ Regularization

80% match2024

A Novel Sparse Regularizer

80% match2023

Learning Sparsity of Representations with Discrete Latent Variables

79% match2023

Cite this paper

@article{botteghi2024sparsifying,
  title  = {Sparsifying Parametric Models with L0 Regularization},
  author = {Nicolò Botteghi and Urban Fasel},
  year   = {2024},
  eprint = {2409.03489},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2409.03489v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2409.03489.svg)](https://codeofpaper.com/paper/2409.03489)

Discussion