~ similar to 2607.03191· 17 results
This paper introduces Mega, a digital architecture for Convolutional Spiking Neural Networks (SNNs) that addresses underutilization of parallelism and inflexibility in existing SNN accelerators throug…
Sven Nitzsche, Alexandru Ionita, Andreas Faust, Bogdan Ionescu +1 more
The paper presents an open-source framework for assessing biological plausibility of spiking neuron models by optimizing parameters to replicate canonical neuronal firing patterns.
Haihang Xia, Xinyu Zhao, Xuecheng Wang, John Goodenough +4 more
This paper proposes and validates a novel hardware architecture, ITP-STDP, to significantly reduce the energy consumption and hardware overhead associated with training Spiking Neural Networks (SNNs).
This paper presents a spiking neural network architecture for generating polar trajectories using a winner-take-all architecture and accessory populations, achieving significant reductions in step tim…
Lingxin Jin, Wei Jiang, Maregu Assefa Habtie, Letian Chen +4 more
The paper introduces Spike-PTSD, a novel, biologically inspired adversarial attack framework that successfully compromises the robustness of Spiking Neural Networks (SNNs) by modeling abnormal neural…
This paper compares three neuromorphic core designs for handling weight sparsity in event-driven neural networks and quantifies the performance and energy costs.
Raj Patel, David Amebley, Taye Akinrele, Shaswata Mitra +2 more
The paper systematically evaluates 27 Spiking Neural Network (SNN) configurations to determine the optimal combination of neuron model and spike encoding scheme for network intrusion detection, findin…
Raj Patel, David Amebley, Taye Akinrele, Shaswata Mitra +2 more
The paper evaluates 27 different Spiking Neural Network (SNN) configurations to determine the optimal design for network intrusion detection, finding that the LeakyParallel neuron combined with latenc…
The paper demonstrates that quadratic integrate-and-fire (QIF) neurons are superior to leaky integrate-and-fire (LIF) neurons for gradient descent training in spiking neural networks because their con…
This paper argues for the importance of modularity and heterogeneity in AI architectures, contrasting the Transformer model with the structure of the cortex.
Turning off automatic NUMA balancing in simulation of large-scale spiking networks reduces energy consumption by 30%.
CLANE presents an end-to-end continual action recognition system deployed on neuromorphic hardware (Intel Loihi 2) using event cameras, achieving high accuracy with massive reductions in energy and la…