# A Deep Neural Network for Finger Counting and Numerosity Estimation

## TL;DR

This paper introduces a deep neural network model that learns finger counting and numerosity estimation, demonstrating the importance of unsupervised pre-training and its similarities to human subitizing behavior.

## Contribution

The paper presents a novel neuro-robotics model combining unsupervised and supervised training for finger counting and number estimation tasks.

## Key findings

- Unsupervised pre-training improves number estimation performance.
- Model shows similarities to human subitizing behavior.
- Extended model can estimate higher numerosities.

## Abstract

In this paper, we present neuro-robotics models with a deep artificial neural network capable of generating finger counting positions and number estimation. We first train the model in an unsupervised manner where each layer is treated as a Restricted Boltzmann Machine or an autoencoder. Such a model is further trained in a supervised way. This type of pre-training is tested on our baseline model and two methods of pre-training are compared. The network is extended to produce finger counting positions. The performance in number estimation of such an extended model is evaluated. We test the hypothesis if the subitizing process can be obtained by one single model used also for estimation of higher numerosities. The results confirm the importance of unsupervised training in our enumeration task and show some similarities to human behaviour in the case of subitizing.