TY - GEN

T1 - REAL-2019

T2 - 33rd Annual Conference on Neural Information Processing Systems, NeurIPS 2019

AU - Cartoni, Emilio

AU - Mannella, Francesco

AU - Santucci, Vieri Giuliano

AU - Triesch, Jochen

AU - Rückert, Elmar

AU - Baldassarre, Gianluca

N1 - Publisher Copyright: © 2020 E. Cartoni, F. Mannella, V.G. Santucci, J. Triesch, E. Rueckert & G. Baldassarre.

PY - 2019

Y1 - 2019

N2 - Open-ended learning, also called ‘life-long learning’ or ‘autonomous curriculum learning’, aims to program machines and robots that autonomously acquire knowledge and skills in a cumulative fashion. We illustrate the first edition of the REAL-2019 – Robot open-Ended Autonomous Learning competition, prompted by the EU project GOAL-Robots – Goal-based Open-ended Autonomous Learning Robots. The competition was based on a simulated robot that: (a) acquires sensorimotor competence to interact with objects on a table; (b) learns autonomously based on mechanisms such as curiosity, intrinsic motivations, and self-generated goals. The competition featured a first ‘intrinsic phase’, where the robots learned to interact with the objects in a fully autonomous way (no rewards, predefined tasks or human guidance), and a second ‘extrinsic phase’, where the acquired knowledge was evaluated with tasks unknown during the first phase. The competition ran online on AIcrowd for six months, involved 75 subscribers and 6 finalists, and was presented at NeurIPS-2019. The competition revealed very hard as it involved difficult machine learning challenges usually tackled in isolation, such as exploration, sparse rewards, object learning, generalisation, catastrophic interference, and autonomous skill learning. Following the participant’s positive feedback, the preparation of a second REAL-2020 competition is underway, improving on the formulation of a relevant benchmark for open-ended learning.

AB - Open-ended learning, also called ‘life-long learning’ or ‘autonomous curriculum learning’, aims to program machines and robots that autonomously acquire knowledge and skills in a cumulative fashion. We illustrate the first edition of the REAL-2019 – Robot open-Ended Autonomous Learning competition, prompted by the EU project GOAL-Robots – Goal-based Open-ended Autonomous Learning Robots. The competition was based on a simulated robot that: (a) acquires sensorimotor competence to interact with objects on a table; (b) learns autonomously based on mechanisms such as curiosity, intrinsic motivations, and self-generated goals. The competition featured a first ‘intrinsic phase’, where the robots learned to interact with the objects in a fully autonomous way (no rewards, predefined tasks or human guidance), and a second ‘extrinsic phase’, where the acquired knowledge was evaluated with tasks unknown during the first phase. The competition ran online on AIcrowd for six months, involved 75 subscribers and 6 finalists, and was presented at NeurIPS-2019. The competition revealed very hard as it involved difficult machine learning challenges usually tackled in isolation, such as exploration, sparse rewards, object learning, generalisation, catastrophic interference, and autonomous skill learning. Following the participant’s positive feedback, the preparation of a second REAL-2020 competition is underway, improving on the formulation of a relevant benchmark for open-ended learning.

KW - autonomous open-ended learning

KW - intrinsic motivations

KW - Simulated robot

UR - http://www.scopus.com/inward/record.url?scp=85162197447&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85162197447

VL - 123.2019

T3 - Proceedings of Machine Learning Research

SP - 142

EP - 152

BT - Proceedings of Machine Learning Research

Y2 - 8 December 2019 through 14 December 2019

ER -