Analogical and Relational Reasoning with Spiking Neural Networks
This addresses the challenge of abstract reasoning in AI, offering a biologically inspired approach that improves generalization and reduces data reliance, though it appears incremental as it builds on existing neural network methods.
The paper tackles abstract reasoning using Raven's Progressive Matrices by augmenting neural networks with spiking modules, achieving human-level accuracy in supervised learning and outperforming existing unsupervised methods.
Raven's Progressive Matrices have been widely used for measuring abstract reasoning and intelligence in humans. However for artificial learning systems, abstract reasoning remains a challenging problem. In this paper we investigate how neural networks augmented with biologically inspired spiking modules gain a significant advantage in solving this problem. To illustrate this, we first investigate the performance of our networks with supervised learning, then with unsupervised learning. Experiments on the RAVEN dataset show that the overall accuracy of our supervised networks surpass human-level performance, while our unsupervised networks significantly outperform existing unsupervised methods. Finally, our results from both supervised and unsupervised learning illustrate that, unlike their non-augmented counterparts, networks with spiking modules are able to extract and encode temporal features without any explicit instruction, do not heavily rely on training data, and generalise more readily to new problems. In summary, the results reported here indicate that artificial neural networks with spiking modules are well suited to solving abstract reasoning.