Logical Information Cells I

TL;DR

This paper investigates how simple artificial neural networks spontaneously develop logical reasoning capabilities, linking neuron activity patterns to semantic information and demonstrating that weights perform logical proofs.

Contribution

It introduces a measure of logical value for neural cells, showing how network depth enhances logical information and that weights encode logical proofs, revealing new insights into neural reasoning.

Findings

Neuron activities become more quantized and informative with depth.

Logical scores correlate with weight sizes, indicating sparsity.

Weights perform logical proofs, enabling classification through logical matrices.

Abstract

In this study we explore the spontaneous apparition of visible intelligible reasoning in simple artificial networks, and we connect this experimental observation with a notion of semantic information. We start with the reproduction of a DNN model of natural neurons in monkeys, studied by Neromyliotis and Moschovakis in 2017 and 2018, to explain how "motor equivalent neurons", coding only for the action of pointing, are supplemented by other neurons for specifying the actor of the action, the eye E, the hand H, or the eye and the hand together EH. There appear inner neurons performing a logical work, making intermediary proposition, for instance E V EH. Then, we remarked that adding a second hidden layer and choosing a symmetric metric for learning, the activities of the neurons become almost quantized and more informative. Using the work of Carnap and Bar-Hillel 1952, we define a measure of the logical value for collections of such cells. The logical score growths with the depth of the layer, i.e. the information on the output decision increases, which confirms a kind of bottleneck principle. Then we study a bit more complex tasks, a priori involving predicate logic. We compare the logic and the measured weights. This shows, for groups of neurons, a neat correlation between the logical score and the size of the weights. It exhibits a form of sparsity between the layers. The most spectacular result concerns the triples which can conclude for all conditions: when applying their weight matrices to their logical matrix, we recover the classification. This shows that weights precisely perform the proofs.