Saved in:
Bibliographic Details
Main Author: Wetzel, Sebastian Johann
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2401.04978
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913524210466816
author Wetzel, Sebastian Johann
author_facet Wetzel, Sebastian Johann
contents I introduce a unified framework for finding a closed-form interpretation of any single neuron in an artificial neural network. Using this framework I demonstrate how to interpret neural network classifiers to reveal closed-form expressions of the concepts encoded in their decision boundaries. In contrast to neural network-based regression, for classification, it is in general impossible to express the neural network in the form of a symbolic equation even if the neural network itself bases its classification on a quantity that can be written as a closed-form equation. The interpretation framework is based on embedding trained neural networks into an equivalence class of functions that encode the same concept. I interpret these neural networks by finding an intersection between the equivalence class and human-readable equations defined by a symbolic search space. The approach is not limited to classifiers or full neural networks and can be applied to arbitrary neurons in hidden layers or latent spaces.
format Preprint
id arxiv_https___arxiv_org_abs_2401_04978
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Closed-Form Interpretation of Neural Network Classifiers with Symbolic Gradients
Wetzel, Sebastian Johann
Machine Learning
Artificial Intelligence
I introduce a unified framework for finding a closed-form interpretation of any single neuron in an artificial neural network. Using this framework I demonstrate how to interpret neural network classifiers to reveal closed-form expressions of the concepts encoded in their decision boundaries. In contrast to neural network-based regression, for classification, it is in general impossible to express the neural network in the form of a symbolic equation even if the neural network itself bases its classification on a quantity that can be written as a closed-form equation. The interpretation framework is based on embedding trained neural networks into an equivalence class of functions that encode the same concept. I interpret these neural networks by finding an intersection between the equivalence class and human-readable equations defined by a symbolic search space. The approach is not limited to classifiers or full neural networks and can be applied to arbitrary neurons in hidden layers or latent spaces.
title Closed-Form Interpretation of Neural Network Classifiers with Symbolic Gradients
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2401.04978