The Foundation for Knowledge Based Systems (SKBS) continued their policy of awarding the SKBS prize to the best demonstration of the Demo-session of the BNAIC 2017.
The 2017 referee committee consisted of Jaap van den Herik (chair, UL, SKBS), Bart Verheij (RUG), Tibor Bosse (vz. BNVKI, VUA, Radboud Universiteit Nijmegen), Annet Onnes (RUG, studentlid), Nico Roos (UM) en Katrien Breuls (VUB).
The referee committee had to consider six submissions which were eligible for the SKBS prize. In Table 1 we list them by topic (in the order of their publication in the Conference Program BNAIC 2017).
|1. Jonathan Gerbscheid, Thomas Groot and Arnoud Visser
Intelligent News Conversation with the Pepper Robot
2. Dennis Steckelmacher, Hélène Plisnier, Diederik M. Roijers and Ann NowéHierarchical Reinforcement Learning for a Robotic Partially Observable Task
3. Peter Vamplew, Dean Webb, Luisa M Zintgraf, Diederik M. Roijers, Richard Dazeley, Rustam Issabekov and Evan Dekker
MORL-Glue: A Benchmark Suite for Multi-Objective Reinforcement Learning
4. Tijn van der Zant and Lars Zwanepol Klinkmeijer
Robocup HQ: A new benchmark focusing on AI, HMI and Autonomous Agents
5. Manon Legrand, Roxana Rădulescu, Diederik M. Roijers and Ann Nowé
The SimuLane Highway Traffic Simulator for Multi-Agent Reinforcement Learning
6. Benjamin Timmermans, Zoltán Szlávik, Manfred Overmeen and Alessandro Bozzon
ECrowd: Enterprise Crowdsourcing for Training Cognitive Systems using the Workforce
Table 1: The 2017 candidates of the SKBS prize.
Since 1999 we have seen many different appearances of the Demo-session. The common characteristic is the emphasis on being “an industrial exhibition”. Up to 2006 the prize money was provided by SKBS only. The Foundation for Knowledge Based Systems originates from the late 1980s as a foundation within SPIN (Stimulerings Projectteam In Nederland). The Foundation SNN (Stichting Neurale Netwerken) is another well-known member of the former SPIN. SNN supported SKBS financially with augmenting the SKBS prize in 2007. In 2008, the industrial partner Strukton announced its willingness to participate in the prize funding. The extra contribution was gratefully accepted. They continued this policy in 2009, 2010, and 2011. Since 2012 SKBS sponsors the BNAIC by Euro 500,- for the best demo Award.
In 2017, six submissions were exhibited in three demonstration rooms for the SKBS prize. All six demos were very interesting, but in fact of a different type. In the early years, say 1999-2006 we usually had 8 to 12 demo’s, but the last ten years the number was between 4 and 8. Obviously, in Amsterdam 2016 we had a heyday for the demo session (with eleven demos). Obviously, over the years, the quality of the contributions has raised. It was a pleasure to see and assess the ingenuity of the demo designers and demo implementations. The jury (referee committee) was given the task to take the following items into consideration by scoring them between 1 and 10: (a) relation to AI, (b) originality, (c) applicability (or is it already a (full-fledged) application?), (d) does it contribute to the further development of AI?, (e) the generalisability to other AI applications/domains, and (f) the contribution to Society (Valorisation).
All in all, the referee committee had a difficult task. The procedure went in shifts: from six we reduced the number of candidates to three and finally to one.
The main observations were as follows. There was a breakthrough of ideas on (1) the robotics, in particular by controlling them via intelligent communication and (2) interaction.
The first prize was assigned to Dennis Steckelmacher, Hélène Plisnier, Diederik M. Roijers and Ann Nowé for the demo Hierarchical Reinforcement Learning for a Robotic Partially Observable Task.
The prize is €500,- and has been awarded to the Robot Team of the Vrije Universiteit Brussels.
In Table 2 we provide an overview of the winners of the SKBS prize so far.
M. van Wezel, J. Sprenger, R. van Stee, and H. La Poutré
Neural Vision 2.0 – Exploratory Data Analysis with Neural Networks
|2000 Kaatsheuvel (shared prize)
Alexander Ypma, Rob Kleiman, Jan Valk, and Bob Duin
MINISOM – A System for Machine Health Monitoring with Neural Networks
F. Brazier, D. Mobach, and B. Overeinder
Bert Kappen, Wim Wiegerinck, Ender Akay, Marcel Nijman, Jan Neijt, and André van Beek
Promedas: A Diagnostic Decision Support System
The Secret Prover: Proving Possession of Arbitrary Files While not Giving Them Away
Gerald de Jong
Fluidiom: The Evolution of Locomotion
Marion Verduijn, Niels Peek, Peter Rosseel, Evert de Jonge, and Bas de Mol
Procarsur: A System for Prognostic Reasoning in Cardiac Surgery
Tim Harbers, Rob van der Veen, Marten den Uyl
Sentient Demonstration BNAIC 07: Vicavision
|2008 Enschede (shared prize)
Joris Maervoet, Patrick De Causmaecker, and Greet Vanden Berghe
A Generic Rule Miner for Geographic Data
Dennis Reidsma and Anton Nijholt
Temporal Interaction between an Artificial Orchestra Conductor and Human Musicians
Tom van Bergen, Maarten Brugmans, Bart Dohmen and Niels Molenaar
Cobes: The clean, safe and hospitable metro
Willem Burgers, Wim Wiegerinck, and Bert Kappen
Disaster Victim Identification System
Wim Vancroonenburg, Jannes Verstichel, Greet Vanden Berghe, and Wouter Souffriau
Efficient aircraft loading: a mixed integer programming approach for the aircraft weight and balance problem
Michel Klein, Nataliya Mogles, and Arlette van Wissen
Demonstration of eMate – Stimulating Behaviour Change via Mobile Phone
Sjriek Alers, Daniel Claes, Joscha Fossel, Daniel Hennes, and Karl Tuyls
Applied Robotics: Precision Placement in RoboCup@Work
Steffen Michels, Marina Velikova, Bas Huijbrechts, Peter Novak, Jesper Hoeksma, Roeland Scheepens, Jan Laarhuis, and André Bonhof
Enhancing Operational Work in Maritime Safety-and-Security Tasks.
Wiebe van Ranst and Joost Vennekes
Ultra-low-latency Endoscopic Image Stabilisation
Caitlin Lagrand, Patrick M. de Kok, Sébastien Negrijn, Michiel van der Meer and Arnoud Visser
Autonomous robot soccer matches
Dennis Steckelmacher, Hélène Plisnier, Diederik M. Roijers and Ann Nowé
Hierarchical Reinforcement Learning for a Robotic Partially Observable Task
Table 2: Overview of SKBS prizes.