Hot Topics in Intelligence Research

Ariana Strandburg-Peshkin (MPI-AB & the University of Konstanz), “Communication and coordination in animal societies”

SCIoI, Marchstraße 23, 10587 Berlin, Room 2.057

Abstract: Many social species use signals such as vocalizations to coordinate a range of group behaviors, from coming to consensus on where to move to banding together against threats. Despite their widespread importance, these behaviors remain challenging to study in the wild because doing so requires monitoring many individuals simultaneously. In this talk, I will

For the Public

Excellent Pub Quiz

Fahimi bar Skalitzer Str. 133, Berlin, Germany

Dive into the wonderful world of research of the seven Berlin Clusters of Excellence: from literature to chemistry, from politics to AI, you and your team can find answers for exciting and surprising questions from the clusters’ research areas. So seek fellows and think of your team name! Every quiz evening focuses on the research

Thursday Morning Talk

Dimitri Coelho Mollo (Umeå University), “Functional Ontologies for AI systems: Tasks, Mechanisms, and Capacities”

SCIoI, Marchstraße 23, 10587 Berlin, Room 2.057

Abstract The size and complexity of current Deep Artificial Neural Networks pose remarkable challenges to our attempts of explaining and understanding their workings. In this talk, I put forward a proposal for complementing existing efforts to that aim, inspired by research on cognitive ontology in philosophy of cognitive science. In particular, I suggest that, as

External Event

Oliver Brock (Science of Intelligence), “Robotics and AI Will Never Feel the Same Again”

Online

The Robotics Institute Germany (RIG) offers weekly insights into current research work on robotics. In RIG's weekly online lectures, leading robotics researchers present the latest developments, challenges, and visions and answer questions from the audience. SCIoI spokesperson Oliver Brock will kick off the series with a talk about the future of robotics and AI. He

Berlin Science Week 2025

Palina Bartashevich, Marten Kaas, Valerii Chirkov, Yunus Sevinchan, and Yating Zheng (Science of Intelligence), “What is Collective Intelligence?”

MAR Asimov Lab Marchstrasse 23, Berlin, Germany

Abstract: From ice fishers and marlins to robots, natural and artificial collectives show remarkable ways of making decisions together. Join us as we explore the collective mind! How do sardines escape the attacks of marlins – and how do marlins adapt their hunting strategies in return? What guides groups of hobby ice fishers in Finland

Berlin Science Week 2025

The Science Slam

Roadrunner's Paradise Saarbrücker Str. 24 10405 Berlin Germany

Amidst the excitement, one event stands out for its humor, heart, and energy: the Battle of the Brains: The Ultimate Science Slam Edition. This much-loved tradition, hosted by Berlin’s seven Clusters of Excellence (SCIoI, MATH+, UniSysCat, SCRIPTS, Neurocure, Temporal Communities, Matters of Activity) transforms complex research into lively and entertaining performances. Researchers bring their work to life with everything from

For the Public

Excellent Pub Quiz

Fahimi bar Skalitzer Str. 133, Berlin, Germany

Dive into the wonderful world of research of the seven Berlin Clusters of Excellence: from literature to chemistry, from politics to AI, you and your team can find answers for exciting and surprising questions from the clusters’ research areas. So seek fellows and think of your team name! Every quiz evening focuses on the research

Thursday Morning Talk

Sole Traverso (Science of Intelligence)

SCIoI, Marchstraße 23, 10587 Berlin, Room 2.057

More details to follow. Image created with DALL-E by Maria Ott.

Guest Lecture

Benjamin Alt (Universität Bremen), “Hybrid AI for Safe Embodied Intelligence”

Marchstraße 23, 10587 Berlin, Room 2.057

Enabling robots to perform complex manipulation tasks in open-world environments requires approaches to intelligence that bridge the traditional divide between symbolic and subsymbolic systems. Hybrid systems that combine reasoning over symbolic knowledge with probabilistically approximately correct inference over probability distributions promise to make robot manipulation actions generalizable, interpretable, and scalable to address real-world challenges. This