NICE 2021 - Agenda
(show all abstracts)Tuesday, 16 March 2021 | |||
NICE #8(The NICE#8 was initially planned to take place in 2020, but had to be postponed to 2021 due to COVID-19.) Note: the exact order and timing of the talks is so far draft! Times in the agenda are in CET (Berlin), EDT (New York), PDT (Los Angeles) and UTC. (Some other time zones: Australia, Japan, China, India, ... or only CET ... ) Meeting venue:
NEUROTECH Forum (15 March 2021)Info: just before NICE on 15 March 2021 the NEUROTECH Forum II will take place online (free of charge. Registration for the forum event). The topic is "Neuromorphic Computing Technologies: Opportunities, challenges and Applications Roadmap". | |||
CET: 14:00‑14:10 EDT: 09:00‑09:10 PDT: 06:00‑06:10 UTC: 13:00‑13:10 (10+5 min) | Welcome to NICE #8 | ||
CET: 14:15‑14:40 EDT: 09:15‑09:40 PDT: 06:15‑06:40 UTC: 13:15‑13:40 (25 min) | Organizer Round | ||
CET: 14:40‑15:20 EDT: 09:40‑10:20 PDT: 06:40‑07:20 UTC: 13:40‑14:20 (40+5 min) | Keynote: Lessons from Loihi for the Future of Neuromorphic Computing | Mike Davies (Intel) | |
CET: 15:25‑15:45 EDT: 10:25‑10:45 PDT: 07:25‑07:45 UTC: 14:25‑14:45 (20+5 min) | Why is Neuromorphic Event-based Engineering the future of AI? | Ryad Benjamin Benosman (UPITT/CMU/SORBONNE) | |
CET: 15:50‑16:10 EDT: 10:50‑11:10 PDT: 07:50‑08:10 UTC: 14:50‑15:10 (20+5 min) | The BrainScaleS mobile platform | Johannes Schemmel (Heidelberg University) | |
CET: 16:15‑16:45 EDT: 11:15‑11:45 PDT: 08:15‑08:45 UTC: 15:15‑15:45 (30 min) | (break) | ||
CET: 16:45‑16:55 EDT: 11:45‑11:55 PDT: 08:45‑08:55 UTC: 15:45‑15:55 (10 min) | Group photo (zoom screenshots) | ||
CET: 16:55‑17:15 EDT: 11:55‑12:15 PDT: 08:55‑09:15 UTC: 15:55‑16:15 (20+5 min) | Evaluating complexity and resilience trade-offs in emerging memory inference machines | Christopher Bennett (Sandia National Labs) | |
CET: 17:20‑17:30 EDT: 12:20‑12:30 PDT: 09:20‑09:30 UTC: 16:20‑16:30 (10+5 min) | Lightning talk: From clean room to machine room: towards accelerated cortical simulations on the BrainScaleS wafer-scale system | Sebastian Schmitt (Heidelberg University) | |
CET: 17:35‑17:55 EDT: 12:35‑12:55 PDT: 09:35‑09:55 UTC: 16:35‑16:55 (20+5 min) | Closed-loop experiments on the BrainScaleS-2 architecture | Korbinian Schreiber (Heidelberg University) | |
CET: 18:00‑18:20 EDT: 13:00‑13:20 PDT: 10:00‑10:20 UTC: 17:00‑17:20 (20+5 min) | Batch << 1: Why Neuromorphic Computing Architectures Suit Real-Time Workloads | Jonathan Tapson (GrAI Matter Labs) | |
CET: 18:25‑18:45 EDT: 13:25‑13:45 PDT: 10:25‑10:45 UTC: 17:25‑17:45 (20+5 min) | Neuromorphic and AI research at BCAI (Bosch Center for Artificial Intelligence) | Thomas Pfeil (Bosch Center for Artificial Intelligence) | |
CET: 18:50‑19:10 EDT: 13:50‑14:10 PDT: 10:50‑11:10 UTC: 17:50‑18:10 (20+5 min) | Mapping Deep Neural Networks on SpiNNaker2 | Florian Kelber (TU Dresden) | |
CET: 19:15‑19:45 EDT: 14:15‑14:45 PDT: 11:15‑11:45 UTC: 18:15‑18:45 (30 min) | Open mic / discussion | ||
CET: 19:45 EDT: 14:45 PDT: 11:45 UTC: 18:45 | End of day I | ||
CET: 19:45‑20:45 EDT: 14:45‑15:45 PDT: 11:45‑12:45 UTC: 18:45‑19:45 (60 min) | (break) | ||
CET: 21:00‑00:00 EDT: 16:00‑19:00 PDT: 13:00‑16:00 UTC: 20:00‑23:00 (180 min) | Tutorials: BrainScaleS and DYNAP-SETwo tutorials/hands on in parallel:
For a description please see the tutorials page. |
Wednesday, 17 March 2021 | |||
CET: 10:30‑12:00 EDT: 05:30‑07:00 PDT: 02:30‑04:00 UTC: 09:30‑11:00 (90 min) | Tutorial: SpiNNaker hands-on(Note: the same SpiNNaker hands on tutorial is also offered on Thursday, 21:00 - 22:30h CET) For a description please see the tutorials page. | Andrew Rowley (UMAN) | |
CET: 12:00‑14:00 EDT: 07:00‑09:00 PDT: 04:00‑06:00 UTC: 11:00‑13:00 (120 min) | (break) | ||
CET: 14:00 EDT: 09:00 PDT: 06:00 UTC: 13:00 | NICE - day II | ||
CET: 14:00‑14:40 EDT: 09:00‑09:40 PDT: 06:00‑06:40 UTC: 13:00‑13:40 (40+5 min) | Keynote: From Brains to Silicon -- Applying lessons from neuroscience to machine learning | Jeff Hawkins and Subutai Ahmad (Numenta) | |
CET: 14:45‑15:05 EDT: 09:45‑10:05 PDT: 06:45‑07:05 UTC: 13:45‑14:05 (20+5 min) | A Neuromorphic Future for Classic Computing Tasks | Brad Aimone (Sandia National Laboratories) | |
CET: 15:10‑15:20 EDT: 10:10‑10:20 PDT: 07:10‑07:20 UTC: 14:10‑14:20 (10+5 min) | Lightning talk: Benchmarking of Neuromorphic Hardware Systems | Christoph Ostrau (Bielefeld University) | |
CET: 15:25‑15:45 EDT: 10:25‑10:45 PDT: 07:25‑07:45 UTC: 14:25‑14:45 (20+5 min) | Natural density cortical models as benchmarks for universal neuromorphic computersThroughout evolution, the cortex has increased in volume from mouse to man by three orders of magnitude, while the architecture at the local scale of a cubic millimeter has largely been conserved in terms of the multi-layered structure and the density of synapses. Furthermore, local cortical networks are similar, independent of whether an area processes visual, auditory, or tactile information. This dual universality raises hope that fundamental principles of cortical computation can be discovered. Although a coherent view of these principles still remains missing, the universality motivated researchers already more than a decade go to start to develop neuromorphic computing systems based on the interaction between neurons by delayed point events and basic parameters of cortical architecture. These systems need to be verified in the sense of accurately representing cortical dynamics and validated in the sense of simulating faster or more energy than software solutions on conventional computers. Such comparisons are only meaningful if they refer to implementations of the same neuronal network model. The role of models changes from mere demonstrations of functionality to quantitative benchmarks. In fields of computer science like computer vision and machine learning the definition of benchmarks helps to quantify progress and drives a constructive competition between research groups. The talk argues that neuromorphic computing needs to advance the development of benchmarks of increasing size and complexity. A model of the cortical microcircuit [1] exemplifies the recent interplay and co-design of alternative hardware architectures enabled by a common benchmark. The model represents neurons with their natural number of synapses and at the same time captures the natural connection probability between neurons in the local volume. Consequently, all questions on the proper scaling of network parameters become irrelevant. The model constitutes a milestone for neuromorphic hardware systems as larger cortical models are necessarily less densely connected. As metrics we discuss the energy consumed per synaptic event and the real-time factor. We illustrate the progress in the past few years and show that a single conventional compute node still keeps up with neuromorphic hardware and achieves sub real-time performance. Finally, the talk exposes the limitations of the microcircuit model as a benchmark and positions cortical multi-area models [2] as a biologically meaningful way of upscaling benchmarks to the next problem size.
This work is partially supported by the European Union's Horizon 2020 (H2020) funding framework under grant agreement no. 945539 (Human Brain Project SGA3) and the Helmholtz Association Initiative and Networking Fund under project number SO-092 (Advanced Computing Architectures, ACA). | Markus Diesmann (Forschungszentrum Jülich GmbH) | |
CET: 15:50‑16:15 EDT: 10:50‑11:15 PDT: 07:50‑08:15 UTC: 14:50‑15:15 (25 min) | Poster Lightning Talks1 min - 1 slide poster appetizers | ||
CET: 16:15‑17:15 EDT: 11:15‑12:15 PDT: 08:15‑09:15 UTC: 15:15‑16:15 (60 min) | Poster session A and coffee | ||
CET: 17:15‑17:35 EDT: 12:15‑12:35 PDT: 09:15‑09:35 UTC: 16:15‑16:35 (20+5 min) | Platform-Agnostic Neural Algorithm Composition using Fugu | William Severa (Sandia National Laboratories) | |
CET: 17:40‑17:50 EDT: 12:40‑12:50 PDT: 09:40‑09:50 UTC: 16:40‑16:50 (10+5 min) | Lightning talk: Implementing Backpropagation for Learning on Neuromorphic Spiking Hardware | Andrew Sornborger (Los Alamos National Laboratory) | |
CET: 17:55‑18:15 EDT: 12:55‑13:15 PDT: 09:55‑10:15 UTC: 16:55‑17:15 (20+5 min) | Inductive bias transfer between brains and machines | Fabian Sinz (University Tübingen) | |
CET: 18:20‑18:30 EDT: 13:20‑13:30 PDT: 10:20‑10:30 UTC: 17:20‑17:30 (10+5 min) | Lightning talk: Spike Latency Reduction generates Efficient Predictive Coding | Pau Vilimelis Aceituno (ETH Zürich) | |
CET: 18:35‑18:45 EDT: 13:35‑13:45 PDT: 10:35‑10:45 UTC: 17:35‑17:45 (10+5 min) | Lightning talk: Cognitive Domain Ontologies: HPCs to Ultra Low Power Neuromorphic Platforms | Chris Yakopcic (University of Dayton) | |
CET: 18:50‑19:20 EDT: 13:50‑14:20 PDT: 10:50‑11:20 UTC: 17:50‑18:20 (30 min) | Open mic / discussion | ||
CET: 19:20‑20:20 EDT: 14:20‑15:20 PDT: 11:20‑12:20 UTC: 18:20‑19:20 (60 min) | (break) |
Thursday, 18 March 2021 | |||
CET: 10:00‑13:00 EDT: 05:00‑08:00 PDT: 02:00‑05:00 UTC: 09:00‑12:00 (180 min) | Tutorial: BrainScaleS hands-on(Note: the same BrainScaleS hands on tutorial is also offered on Tuesday evening)
For a description of the pre-requirements, please see the tutorials page. | ||
CET: 13:00‑14:00 EDT: 08:00‑09:00 PDT: 05:00‑06:00 UTC: 12:00‑13:00 (60 min) | (break) | ||
CET: 14:00 EDT: 09:00 PDT: 06:00 UTC: 13:00 | NICE - day III | ||
CET: 14:00‑14:40 EDT: 09:00‑09:40 PDT: 06:00‑06:40 UTC: 13:00‑13:40 (40+5 min) | Keynote: Biological inspiration for improving computing and learning in spiking neural networks | Wolfgang Maass (Graz University of Technology) | |
CET: 14:45‑15:05 EDT: 09:45‑10:05 PDT: 06:45‑07:05 UTC: 13:45‑14:05 (20+5 min) | On the computational power and complexity of Spiking Neural Networks | Johan Kwisthout (Radboud Universiteit Nijmegen) | |
CET: 15:10‑15:30 EDT: 10:10‑10:30 PDT: 07:10‑07:30 UTC: 14:10‑14:30 (20+5 min) | Evolutionary Optimization for Neuromorphic Systems | Catherine Schuman (Oak Ridge ) | |
CET: 15:35‑15:55 EDT: 10:35‑10:55 PDT: 07:35‑07:55 UTC: 14:35‑14:55 (20+5 min) | An event-based gas sensing device that resolves fast transients in a turbulent environment | Michael Schmuker (University of Hertfordshire) | |
CET: 16:00‑16:20 EDT: 11:00‑11:20 PDT: 08:00‑08:20 UTC: 15:00‑15:20 (20+5 min) | Sequence learning, prediction, and generation in networks of spiking neurons | Younes Bouhadjar (Forschungszentrum Juelich) | |
CET: 16:25‑17:25 EDT: 11:25‑12:25 PDT: 08:25‑09:25 UTC: 15:25‑16:25 (60 min) | Poster session b and coffee | ||
CET: 17:25‑17:45 EDT: 12:25‑12:45 PDT: 09:25‑09:45 UTC: 16:25‑16:45 (20+5 min) | Walter Senn (Universität Bern) | ||
CET: 17:50‑18:10 EDT: 12:50‑13:10 PDT: 09:50‑10:10 UTC: 16:50‑17:10 (20+5 min) | Conductance-based dendrites perform reliability-weighted opinion pooling | Jakob Jordan (Institute of Physiology, University of Bern) | |
CET: 18:15‑18:25 EDT: 13:15‑13:25 PDT: 10:15‑10:25 UTC: 17:15‑17:25 (10+5 min) | Lightning talk: Natural gradient learning for spiking neurons | Elena Kreutzer (University of Bern) | |
CET: 18:30‑18:50 EDT: 13:30‑13:50 PDT: 10:30‑10:50 UTC: 17:30‑17:50 (20+5 min) | Making spiking neurons more succinct with multi-compartment models | Johannes Leugering (Fraunhofer IIS) | |
CET: 18:55‑19:05 EDT: 13:55‑14:05 PDT: 10:55‑11:05 UTC: 17:55‑18:05 (10+5 min) | Lightning talk: The Computational Capacity of Mem-LRC Reservoirs | Forrest Sheldon (Los Alamos National Lab - T-4/CNLS) | |
CET: 19:10‑19:40 EDT: 14:10‑14:40 PDT: 11:10‑11:40 UTC: 18:10‑18:40 (30 min) | Open mic / discussion | ||
CET: 19:40‑20:50 EDT: 14:40‑15:50 PDT: 11:40‑12:50 UTC: 18:40‑19:50 (70 min) | (break) | ||
CET: 21:00‑22:30 EDT: 16:00‑17:30 PDT: 13:00‑14:30 UTC: 20:00‑21:30 (90 min) | Tutorial: SpiNNaker hands-on(Note: the same SpiNNaker hands on tutorial is also offered on Wednesday, 10:30-12:00h CET) For a description please see the tutorials page. |