Explaining high-dimensional text classifiers

2023/11/22 by Melamed, Odelia, Caruana, Rich
#Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Neural and Evolutionary Computing (cs.NE)

paper · doi:10.48550/arxiv.2311.13454

Abstract

Explainability has become a valuable tool in the last few years, helping humans better understand AI-guided decisions. However, the classic explainability tools are sometimes quite limited when considering high-dimensional inputs and neural network classifiers. We present a new explainability method using theoretically proven high-dimensional properties in neural network classifiers. We present two usages of it: 1) On the classical sentiment analysis task for the IMDB reviews dataset, and 2) our Malware-Detection task for our PowerShell scripts dataset.

Related

• GRACE: Generating Concise and Informative Contrastive Sample to Explain Neural Network Model's Prediction
• A Diagnostic Study of Explainability Techniques for Text Classification
• Simplifying the explanation of deep neural networks with sufficient and necessary feature-sets: case of text classification
• Towards Human-Centred Explainability Benchmarks For Text Classification
• Human-grounded Evaluations of Explanation Methods for Text\n Classification