FI-CBL: A Probabilistic Method for Concept-Based Learning with Expert Rules

TL;DR

FI-CBL is a probabilistic approach for concept-based learning that integrates expert rules, improves performance with limited data, and offers transparency through explicit probabilistic calculations.

Contribution

It introduces FI-CBL, a novel method combining frequentist inference with expert rule integration for concept-based learning.

Findings

Outperforms concept bottleneck model with small training data

Incorporates expert rules into probabilistic inference

Provides transparent and interpretable calculations

Abstract

A method for solving concept-based learning (CBL) problem is proposed. The main idea behind the method is to divide each concept-annotated image into patches, to transform the patches into embeddings by using an autoencoder, and to cluster the embeddings assuming that each cluster will mainly contain embeddings of patches with certain concepts. To find concepts of a new image, the method implements the frequentist inference by computing prior and posterior probabilities of concepts based on rates of patches from images with certain values of the concepts. Therefore, the proposed method is called the Frequentist Inference CBL (FI-CBL). FI-CBL allows us to incorporate the expert rules in the form of logic functions into the inference procedure. An idea behind the incorporation is to update prior and conditional probabilities of concepts to satisfy the rules. The method is transparent because it has an explicit sequence of probabilistic calculations and a clear frequency interpretation. Numerical experiments show that FI-CBL outperforms the concept bottleneck model in cases when the number of training data is small. The code of proposed algorithms is publicly available.