Year: 2017
Type: conference-paper
Abstract: Recently, a new class of machine learning algorithms has emerged, where models \nand discriminators are generated in a competitive setting. The most prominent \nexample is Generative Adversarial Networks (GANs). In this paper we examine \nhow these algorithms relate to the famous Turing test, and derive what—from a \nTuring perspective—can be considered their defining features. Based on these \nfeatures, we outline directions for generalizing GANs—resulting in the family \nof algorithms referred to as Turing Learning. One such direction is to allow the \ndiscriminators to interact with the processes from which the data samples are \nobtained, making them “interrogators”, as in the Turing test. We validate this idea \nusing two case studies. In the first case study, a computer infers the behavior of an \nagent while controlling its environment. In the second case study, a robot infers \nits own sensor configuration while controlling its movements. The results confirm \nthat by allowing discriminators to interrogate, the accuracy of models is improved.
Authors: Roderich Groß, Yue Gu, Wei Li, Melvin Gauci
Institutions: University of Sheffield, Aston University, Harvard University
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2753347993