Imperial College London
Alternate

ProfessorMurrayShanahan

Faculty of Engineering, Department of Computing

Professor in Cognitive Robotics
 
 
 
//

Contact

 

+44 (0)20 7594 8262m.shanahan Website

 
 
//

Location

 

407BHuxley BuildingSouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@unpublished{Beyret:2019,
author = {Beyret, B and Hernández-Orallo, J and Cheke, L and Halina, M and Shanahan, M and Crosby, M},
title = {The Animal-AI Environment: Training and Testing Animal-Like Artificial  Cognition},
url = {http://arxiv.org/abs/1909.07483v2},
year = {2019}
}
Download

RIS format (EndNote, RefMan)

TY  - UNPB
AB  - Recent advances in artificial intelligence have been strongly driven by theuse of game environments for training and evaluating agents. Games are oftenaccessible and versatile, with well-defined state-transitions and goalsallowing for intensive training and experimentation. However, agents trained ina particular environment are usually tested on the same or slightly varieddistributions, and solutions do not necessarily imply any understanding. If wewant AI systems that can model and understand their environment, we needenvironments that explicitly test for this. Inspired by the extensiveliterature on animal cognition, we present an environment that keeps all thepositive elements of standard gaming environments, but is explicitly designedfor the testing of animal-like artificial cognition.
AU  - Beyret,B
AU  - Hernández-Orallo,J
AU  - Cheke,L
AU  - Halina,M
AU  - Shanahan,M
AU  - Crosby,M
PY  - 2019///
TI  - The Animal-AI Environment: Training and Testing Animal-Like Artificial  Cognition
UR  - http://arxiv.org/abs/1909.07483v2
ER  -
Download