Auxiliary Learning as a step towards Artificial General Intelligence

TL;DR

This paper explores Auxiliary Learning, a method to enhance neural networks' ability to recognize unknown objects, aiming to move closer to Artificial General Intelligence by increasing model generality.

Contribution

It introduces Auxiliary Learning with an auxiliary class to improve neural networks' recognition of unknown objects and generality.

Findings

Auxiliary Learning improves recognition of unseen objects.

Using auxiliary classes enhances neural network generality.

The approach demonstrates potential for advancing towards AGI.

Abstract

Auxiliary Learning is a machine learning approach in which the model acknowledges the existence of objects that do not come under any of its learned categories.The name Auxiliary learning was chosen due to the introduction of an auxiliary class. The paper focuses on increasing the generality of existing narrow purpose neural networks and also highlights the need to handle unknown objects. The Cat & Dog binary classifier is taken as an example throughout the paper.