Mitigating Catastrophic Forgetting in Time-To-First-Spike Neural Networks through Active Dendrites
Active dendrites enable efficient continual learning in time-to-first-spike neural networks, mitigating catastrophic forgetting.
Slax: Eine modulare JAX-Bibliothek für schnelles und flexibles Prototyping von Spiking Neural Networks
Slax ist eine JAX-basierte Bibliothek, die das Design und die Implementierung vielfältiger Trainingsalgorithmen für Spiking Neural Networks beschleunigt und vereinfacht, um die Leistung, Effizienz und Flexibilität von SNNs zu verbessern.
Slax: A Flexible and Efficient JAX Library for Rapid Prototyping of Spiking Neural Networks
Slax is a JAX-based library designed to facilitate the exploration and implementation of diverse training algorithms for spiking neural networks, with a strong emphasis on flexibility and efficiency.
Approximation and Learning Capabilities of Affine Spiking Neural Networks
Affine spiking neural networks, which have positive synaptic weights and are equipped with affine encoders and decoders, can approximate a wide range of functions at optimal rates while exhibiting superior generalization performance compared to feedforward neural networks.
SpikingJET: A Comprehensive Fault Injection Framework for Evaluating the Resilience of Spiking Neural Networks
SpikingJET is a novel fault injector designed specifically for fully connected and convolutional spiking neural networks, enabling comprehensive assessment of their resilience to hardware faults.
Exploring Spiking Lottery Tickets for Efficient and Sparse Spiking Neural Networks
Spiking lottery tickets can achieve higher sparsity and less performance degradation compared to standard lottery tickets, enabling the development of efficient and energy-saving spiking neural networks.
© 2024 by Linnk AI