A three day in-person symposium from Nov 27th to Nov 29th in Mainz, Germany. Organized by Simone Schaub-Meyer & Martin Mundt, enabled by the Hessian Center for Artificial Intelligence – hessian.AI – Connectom network funding program.
Abstract: Continual, or lifelong, machine learning addresses crucial questions that arise when aiming to overcome the limitations of single training cycles, rigid inference engines, and fixed datasets. At the center, it acknowledges that data selection, models, training algorithms, and evaluation measures are not static. Despite its recently emerged popularity, current research has just started to grasp how we can accommodate these factors into human-like lifelong learning AI systems. Although presently ongoing efforts start to take into account complex sequences of datasets, they focus predominantly on image classification tasks. Unfortunately, such tasks significantly simplify learning, e.g. by only performing image-level recognition, assuming that all data is always labelled and that continual learning consists mainly of learning to recognize new object types. In this initiative, we thus set out to bring together researchers who are at the forefront of continual learning and computer vision to jointly catalyze the foundation for continual learning beyond classification. In an inaugural symposium, we will establish common grounds and discover synergies towards continual learning for a plethora of relevant computer vision tasks, such as semantic segmentation and learning without labels.
Symposium Schedule
Sunday 26th
- 15:00 Hotel check-in opens
- 16:00 Simone Schaub-Meyer & Martin Mundt – TU Darmstadt & hessian.AI
Brainstorming with participants on the expectations, individual goals, and potential outcomes of the symposium - 18:00 Dinner
Monday 27th
- 9:00 Simone Schaub-Meyer & Martin Mundt – TU Darmstadt & hessian.AI
Introduction to the organizers’ groups and their respective research - 10:00 Coffee break
- 10:30 Subarnaduti Paul – TU Darmstadt & hessian.AI
Learning Continually from Indirect Experience in Distributed Setups - 11:15 Vincenzo Lomonaco – University of Pisa
Continual, Decentralized Compositionality for Collective Machine Extelligence - 12:00 Lunch break
- 13:30 Francesca Pistilli & Antonio Alliegro – Politecnico di Torino
Advancing Real-World Reasoning Beyond Classification - 14:15 Fabian Hinder – University of Bielefeld
Understanding Concept Drift: From Explanation to Application - 15:00 Coffee break
- 15:30 Discussion
- 17:30 Break and later dinner
Tuesday 28th
- 9:00 Raffaello Camoriano – Politecnico di Torino
Bridging Continual Learning & Structured Prediction for Embodied Systems: Challenges & Open Problems - 9:45 Tinne Tuytelaars – KU Leuven
Continual learning in representations and diffusion models - 10:30 Coffee break
- 11:00 Discussion
- 12:00 Lunch break
- 13:30 Nikhil Churamani – Cambridge University
Continual Learning for Embodied Social Artificial Intelligence - 14:15 Carlo D’Eramo & Georgia Chalvatzaki – University of Würzburg/hessian.AI & TU Darmstadt/hessian.AI
Curriculum Reinforcement Learning for Efficiency and Robustness - 15:00 Coffee break
- 15:30 Discussion
- 17:30 Break and later dinner
Wednesday 29th
- 9:00 Rupert Mitchell – TU Darmstadt & hessian.AI
Self Expanding Neural Networks: a Sustainable & Adaptive Model Foundation - 9:45 Coffee break
- 10:15 Discussion
- 11:45 Simone Schaub-Meyer & Martin Mundt – TU Darmstadt & hessian.AI
Wrap-up by the organizers
Location & Travel
SyCLeC will take place at Gutenberg Digital Hub e.V. in the immediate vicinity of the H2 Hotel near the beautiful Rhine river in Mainz, Germany.
Both are about 2 kilometers away from the Mainz main station, a nice brisk walk should the late November decide to be nice. If the weather is poor or you prefer to take a bus to cut down the walk, then you can take line 9 to Kaisertor Mainz or line 76 to Mainz Feldbergpl. (near the Kunsthalle Mainz).
If you are arriving at Frankfurt Airport, then Mainz main station is but a 25 minute journey away by train. There exist several options, but the easiest is likely to hop onto one of the frequently departing S8 trains in the direction of Wiesbaden Hauptbahnhof.