DT-SCNN:双门尖卷积神经网络,具有较少的操作和边缘应用程序的内存访问
Fuming Lei1,2, Xu Yang1, Jian Liu1,2
1State Key Laboratory of Superlattices and Microstructures, Institute of Semiconductors, Chinese Academy of Sciences, Beijing, China.
Frontiers in computational neuroscience
|June 14, 2024
概括
一个新的双值尖端卷积神经网络 (DT-SCNN) 减少了尖端神经网络 (SNN) 的计算负载和内存使用. 这种方法可以在边缘设备上实现高效的视觉处理,精度损失最小.
科学领域:
- 神经形态工程的神经形态工程
- 人工智能的人工智能
- 计算机视觉 计算机视觉
背景情况:
- 尖端卷积神经网络 (SCNN) 为神经形态硬件上的视觉任务提供高精度和功率效率,使其适合边缘应用.
- 在资源有限的边缘设备上实施SCNNs是具有挑战性的,因为它需要大量的卷积操作和膜电位 (Vm) 存储要求.
- 现有的优化方法,如时间步骤减小,修剪和量化,忽视了尖端特征图 (SFmaps) 中的冗余性.
研究的目的:
- 引入一个双门尖卷积神经网络 (DT-SCNN),通过利用SFmaps中的相似性来减少计算操作和内存访问.
- 通过将Vm和卷积重量存储减半来减少操作数量和内存足迹.
主要方法:
- 拟议的DT-SCNN使用双点火值来从单个Vm地图中生成两个相似的SFmap.
- 使用一种修改的时空反向传播 (STBP) 训练方法,采用两阶段的策略来训练DT-SCNN进行单阶段时间推理.
- 该方法侧重于减少SFmaps中的冗余,以优化SCNN实施.
主要成果:
- 与标准SCNN相比,DT-SCNN实现了对卷积层的操作和数据存储减少50%.
- 精度损失是最小的,在CIFAR10,MNIST和时尚MNIST数据集上不超过0.4%.
- 由于其轻量级的设计和单个时间步骤的推断,DT-SCNN在以前的作品中显示了最低的操作数量.
结论:
- DT-SCNN有效地解决了在边缘设备上实施SCNN的计算和内存限制.
- 双门机制和优化培训显著降低了资源需求,而不会影响准确性.
- 这项工作为为边缘计算环境开发低延迟,节能AI应用铺平了道路.
相关概念视频
01:19The Role of Ion Channels in Neuronal Computation
3.2K
A postsynaptic neuron usually receives numerous impulses from several other presynaptic neurons. The axon hillock of the postsynaptic neuron integrates all these signals and determines the likelihood of firing an action potential.
Sometimes a single EPSP is strong enough to induce an action potential in the postsynaptic neuron. However, multiple presynaptic inputs must often create EPSPs around the same time for the postsynaptic neuron to be sufficiently depolarized to fire an action potential....
Sometimes a single EPSP is strong enough to induce an action potential in the postsynaptic neuron. However, multiple presynaptic inputs must often create EPSPs around the same time for the postsynaptic neuron to be sufficiently depolarized to fire an action potential....
3.2K
01:25Neural Circuits
1.1K
Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
1.1K