<div dir="ltr">Dear Colleagues,<br><br>The GOAL-Robots consortium (<a href="http://www.goal-robots.eu/" target="_blank">http://www.goal-robots.eu/</a>)
is glad to announce the REAL (Robot open-Ended Autonomous Learning)
<span class="gmail-m_-1304844593257386684gmail-il">competition</span>, which will be part of the NeurIPS 2019 <span class="gmail-m_-1304844593257386684gmail-il">competition</span> track (<a href="https://nips.cc/Conferences/2019/CompetitionTrack" target="_blank">https://nips.cc/Conferences/2019/CompetitionTrack</a>). <br><br>The <span class="gmail-m_-1304844593257386684gmail-il">competition</span> addresses the topic of autonomous open-ended learning, and in particular is focused on systems that:<br>(a) acquire sensorimotor competence that allows them to interact with objects and physical environments;<br>(b)
learn in a fully autonomous way, i.e. with no human intervention, on
the basis of mechanisms such as curiosity, intrinsic motivations,
task-free reinforcement learning, self-generated goals, and any
other mechanism that might support autonomous learning.<br><br>The objective of REAL is to:<br>(a) track the state-of-the-art in robot open-ended autonomous learning; <br>(b) foster research and the proposal of new solutions to the many problems posed by open-ended learning;<br>(c) favour the development of benchmarks in the field<br><br>The
systems submitted to the <span class="gmail-m_-1304844593257386684gmail-il">competition</span> will undergo a two-phase life
where during a first ‘intrinsic phase’ the system will have a certain
time to freely explore and learn in the environment, and then during an
`extrinsic phase’ the quality of the autonomously acquired knowledge
will be measured with multiple tasks.<br><br><div>The <span class="gmail-m_-1304844593257386684gmail-il">competition</span> is based on
a PyBullet fast simulator that can be easily installed on own computer
through the software Starting Kit (<a href="https://github.com/GOAL-Robots/REALCompetitionStartingKit" target="_blank">https://github.com/GOAL-Robots/REALCompetitionStartingKit</a>). The simulator involves a robotic arm
with gripper and a camera, and a working table containing objects to
manipulate.</div><br>Submissions will be open from August the 7th.
For further and more specific details please refer to the <span class="gmail-m_-1304844593257386684gmail-il">competition</span>
website (<a href="https://www.aicrowd.com/challenges/robot-open-ended-autonomous-learning-real" target="_blank">https://www.aicrowd.com/challenges/robot-open-ended-autonomous-learning-real</a>).<div><br></div><div>The following contacts are available for further information:<br>Emilio Cartoni <a href="mailto:emilio.cartoni@istc.cnr.it" target="_blank">emilio.cartoni@istc.cnr.it</a><br><div>Sharada Mohanty <a href="mailto:mohanty@aicrowd.com" target="_blank">mohanty@aicrowd.com</a><div><span class="gmail-m_-1304844593257386684gmail-m_7996784301348867801gmail-il">Gianluca</span> Baldassarre <a href="mailto:gianluca.baldassarre@istc.cnr.it" target="_blank"><span class="gmail-m_-1304844593257386684gmail-m_7996784301348867801gmail-il">gianluca</span>.baldassarre@istc.cnr.it</a></div><div><br>Best regards,<br>Vieri Giuliano Santucci</div></div></div><br clear="all"><br>-- <br><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div>Vieri Giuliano Santucci<br><br>Laboratory of Computational Embodied Neuroscience (LOCEN),<br>
Istituto di Scienze e Tecnologie della Cognizione (ISTC),<br>
Consiglio Nazionale delle Ricerche (CNR),<br>
Via San Martino della Battaglia 44, 00185 Roma, Italy<br></div></div></div></div>