Výsledky vyhledávání

Report

Reconsidering the energy efficiency of spiking neural networks

Autor: Yan, Zhanglu, Bai, Zhenyu, Wong, Weng-Fai

Spiking neural networks (SNNs) are generally regarded as more energy-efficient because they do not use multiplications. However, most SNN works only consider the counting of additions to evaluate energy consumption, neglecting other overheads such as

Externí odkaz: http://arxiv.org/abs/2409.08290

Zobrazit plný text záznamu

Report

SWAT: Scalable and Efficient Window Attention-based Transformers Acceleration on FPGAs

Autor: Bai, Zhenyu, Dangi, Pranav, Li, Huize, Mitra, Tulika

Efficiently supporting long context length is crucial for Transformer models. The quadratic complexity of the self-attention computation plagues traditional Transformers. Sliding window-based static sparse attention mitigates the problem by limiting

Externí odkaz: http://arxiv.org/abs/2405.17025

Zobrazit plný text záznamu

Report

SparrowSNN: A Hardware/software Co-design for Energy Efficient ECG Classification

Autor: Yan, Zhanglu, Bai, Zhenyu, Mitra, Tulika, Wong, Weng-Fai

Heart disease is one of the leading causes of death worldwide. Given its high risk and often asymptomatic nature, real-time continuous monitoring is essential. Unlike traditional artificial neural networks (ANNs), spiking neural networks (SNNs) are w

Externí odkaz: http://arxiv.org/abs/2406.06543

Zobrazit plný text záznamu

Report

Computing Execution Times with eXecution Decision Diagrams in the Presence of Out-Of-Order Resources

Autor: Bai, Zhenyu, Cassé, Hugues, Carle, Thomas, Rochange, Christine

Worst-Case Execution Time (WCET) is a key component for the verification of critical real-time applications. Yet, even the simplest microprocessors implement pipelines with concurrently-accessed resources, such as the memory bus shared by fetch and m

Externí odkaz: http://arxiv.org/abs/2207.07481

Zobrazit plný text záznamu