# Composing Neural Learning and Symbolic Reasoning with an Application to Visual Discrimination

## TL;DR

This paper introduces a neurosymbolic framework for visual discrimination puzzles that combines neural networks with symbolic reasoning to produce interpretable classifiers, outperforming purely neural methods.

## Contribution

It presents a novel compositional neurosymbolic approach for solving visual discrimination puzzles with interpretable outputs, bridging neural perception and symbolic reasoning.

## Key findings

- Framework performs well on diverse VDP datasets
- Neurosymbolic approach yields interpretable classifiers
- Outperforms several neural-only methods

## Abstract

We consider the problem of combining machine learning models to perform higher-level cognitive tasks with clear specifications. We propose the novel problem of Visual Discrimination Puzzles (VDP) that requires finding interpretable discriminators that classify images according to a logical specification. Humans can solve these puzzles with ease and they give robust, verifiable, and interpretable discriminators as answers. We propose a compositional neurosymbolic framework that combines a neural network to detect objects and relationships with a symbolic learner that finds interpretable discriminators. We create large classes of VDP datasets involving natural and artificial images and show that our neurosymbolic framework performs favorably compared to several purely neural approaches.