Loading Now

Summary of Sparse Explanations Of Neural Networks Using Pruned Layer-wise Relevance Propagation, by Paulo Yanez Sarmiento et al.


Sparse Explanations of Neural Networks Using Pruned Layer-Wise Relevance Propagation

by Paulo Yanez Sarmiento, Simon Witzke, Nadja Klein, Bernhard Y. Renard

First submitted to arxiv on: 22 Apr 2024

Categories

  • Main: Machine Learning (cs.LG)
  • Secondary: None

     Abstract of paper      PDF of paper


GrooveSquid.com Paper Summaries

GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!

Summary difficulty Written by Summary
High Paper authors High Difficulty Summary
Read the original abstract here
Medium GrooveSquid.com (original content) Medium Difficulty Summary
The proposed approach modifies layer-wise relevance propagation (LRP) to improve explainability in deep neural networks (DNNs). By directly pruning the relevance propagation for different layers, the method achieves sparser attributions for input features and intermediate layers. This allows for pruning different neurons for different inputs, making it more suitable for local explanation methods. The approach is evaluated on image and genomic sequence data, demonstrating reduced noise and concentrated relevance on important features compared to the baseline.
Low GrooveSquid.com (original content) Low Difficulty Summary
The paper talks about making deep neural networks (DNNs) easier to understand. Right now, people need to figure out what’s important in a DNN’s output, which can be hard, especially when dealing with complex data like genome sequences. The researchers suggest modifying an existing explanation method called layer-wise relevance propagation to make it more useful for real-world applications. They test their approach on images and genomic sequences, showing that it reduces noise and helps identify the most important features.

Keywords

» Artificial intelligence  » Pruning