The goal of this workshop was to bring together experts in Theoretical Computer Science and Machine Learning and to discuss recent developments on Algorithmic Aspects of Neural Networks and related topics. The workshop took place on February 5th/6th in the main university building.
Organizer:
- Christian Sohler (University of Cologne)
Schedule:
Monday, 5th February
| 10:00-10:50 | Registration |
| 10:50-11:00 | Welcome |
| 11:00-11:30 | Sitan Chen (Harvard University) - Provably learning a multi-head attention layer |
| 11:30-12:00 | Amir Zandieh (Independent Scientist) - Long-Context Attention in Near-Linear Time |
| 12:00-12:30 | Varun Kanade (University of Oxford) - In-context learning: A case study with learning discrete functions |
| 12:30- 14:00 | Lunch break |
| 14:00-14:30 | Vincent Cohen-Addad (Google Research) - Sensitivity Sampling for Coreset-Based Data Selection |
| 14:30-15:00 | Dan Feldman (University of Haifa) - Coresets for network compression |
| 15:00-15:30 | Coffee break |
| 15:30-16:00 | Silvio Lattanzi (Google Research) - Clustering with ML Advice |
| 16:00-16:30 | Piotr Indyk (MIT) - A Near-Linear Time Algorithm for the Chamfer Distance |
| 16:30-17:00 | Coffee Break |
| 17:00-18:00 | Open Problems and discussions |
Tuesday, 6th February
| 09:00-09:30 | Sandra Kiefer (University of Oxford) -The Combinatorial Side of Graph Neural Networks |
| 09:30-10:00 | Christopher Morris (RWTH Aachen) - Generalization Abilities of Graph Neural Networks |
| 10:00-10:30 | Aleksandar Bojchevski (University of Cologne) - Machine Learning with Guarantees |
| 10:30-11:00 | Coffee Break |
| 11:00-11:30 | Michael Kapralov (EPFL) - A Quasi-Monte Carlo Data Structure for Smooth Kernel Evaluation |
| 11:30-12:30 | Goran Zuzic (Google Research) - A Simple Boosting Framework for Transshipment |
| 12:30-14:00 | Lunch break |
| 14:00-14:30 | Vladimir Braverman (Rice University) - FetchSGD: Communication-Efficient Federated Learning with Sketching |
| 14:30-15:00 | Ameya Velingker (Google Research) - Oversquashing in Graph Neural Networks |
| 15:00-15:30 | Alexander Munteanu (TU Dortmund) - Bounding the Width of Neural Networks |
| 15:30-16:00 | Coffee Break |
| 16:00-16:30 | David Woodruff (CMU) - On Approximating an Entrywise Transformed Matrix Product |
| 16:30-17:00 | Artur Czumaj (University of Warwick) - Testing Neural Networks |
| 17:00-18:00 | Open Problems and discussions |