A5051518153- Advanced Memory and Neural Computing
- Neural dynamics and brain function
- Ferroelectric and Negative Capacitance Devices
- Neural Networks and Reservoir Computing
- Neural Networks and Applications
- Functional Brain Connectivity Studies
- Infrastructure Maintenance and Monitoring
- Advanced Neural Network Applications
- CCD and CMOS Imaging Sensors
- Industrial Vision Systems and Defect Detection
- Non-Destructive Testing Techniques
Kirchhoff (Germany)
2019-2025
Heidelberg University
2019-2025
University of Bern
2019-2025
Institute for Physics
2019-2020
We present first experimental results on the novel BrainScaleS-2 neuromorphic architecture based an analog neuro-synaptic core and augmented by embedded microprocessors for complex plasticity experiment control. The high acceleration factor of 1000 compared to biological dynamics enables execution computationally expensive tasks, allowing fast emulation long-duration experiments or rapid iteration over many consecutive trials. flexibility our is demonstrated in a suite five distinct...
Neuromorphic systems open up opportunities to enlarge the explorative space for computational research. However, it is often challenging unite efficiency and usability. This work presents software aspects of this endeavor BrainScaleS-2 system, a hybrid accelerated neuromorphic hardware architecture based on physical modeling. We introduce key Operating System: experiment workflow, API layering, design, platform operation. present use cases discuss derive requirements showcase implementation....
For a biological agent operating under environmental pressure, energy consumption and reaction times are of critical importance. Similarly, engineered systems also strive for short time-to-solution low energy-to-solution characteristics. At the level neuronal implementation, this implies achieving desired results with as few early spikes possible. In time-to-first-spike coding framework, both these goals inherently emerging features learning. Here, we describe rigorous derivation...
The Yin-Yang dataset was developed for research on biologically plausible error backpropagation and deep learning in spiking neural networks. It serves as an alternative to classic datasets, especially early-stage prototyping scenarios both network models hardware platforms, which it provides several advantages. First, is smaller therefore faster learn, thereby being better suited small-scale exploratory studies software simulations prototypes. Second, exhibits a very clear gap between the...
The Yin-Yang dataset was developed for research on biologically plausible error backpropagation and deep learning in spiking neural networks. It serves as an alternative to classic datasets, especially early-stage prototyping scenarios both network models hardware platforms, which it provides several advantages. First, is smaller therefore faster learn, thereby being better suited small-scale exploratory studies software simulations prototypes. Second, exhibits a very clear gap between the...
For a biological agent operating under environmental pressure, energy consumption and reaction times are of critical importance. Similarly, engineered systems optimized for short time-to-solution low energy-to-solution characteristics. At the level neuronal implementation, this implies achieving desired results with as few early spikes possible. With time-to-first-spike coding both these goals inherently emerging features learning. Here, we describe rigorous derivation learning rule such...
With an increasing presence of science throughout all parts society, there is a rising expectation for researchers to effectively communicate their work and, equally, teachers discuss contemporary findings in classrooms. While the community can resort established set teaching aids fundamental concepts most natural sciences, need similarly illustrative experiments and demonstrators neuroscience. We therefore introduce Lu.i: parametrizable electronic implementation leaky-integrate-and-fire...
Recent efforts have fostered significant progress towards deep learning in spiking networks, both theoretical and silico. Here, we discuss several different approaches, including a tentative comparison of the results on BrainScaleS-2, hint future such comparative studies.
We demonstrate five experiments that are performed on the novel BrainScaleS-2 neuromorphic architecture based an analog neuro-synaptic core and augmented by embedded microprocessors for complex plasticity experiment control. The high acceleration factor of 1000 compared to biological dynamics enables execution computationally expensive tasks, allowing fast emulation long-duration or rapid iteration over many consecutive trials. were chosen emphasize different aspects system, especially...
Neuromorphic systems open up opportunities to enlarge the explorative space for computational research. However, it is often challenging unite efficiency and usability. This work presents software aspects of this endeavor BrainScaleS-2 system, a hybrid accelerated neuromorphic hardware architecture based on physical modeling. We introduce key Operating System: experiment workflow, API layering, design, platform operation. present use cases discuss derive requirements showcase implementation....