可配置深度神经网络实现能量-精度权衡
Runtime Configurable Deep Neural Networks for Energy-Accuracy Trade-off
摘要:
我们提出了一种新的动态配置技术,深度神经网络,允许逐步能源精度权衡在运行时。我们的配置技术根据响应时间、功率和精度目标动态调整网络中的通道数量。为了实现这种动态配置技术,我们共同设计了一种新的训练算法,其中网络是递增训练的,使得在早期步骤中训练的通道中的权重是固定的。我们的技术提供了多个网络的灵活性,同时存储和利用一组权重。我们使用基于ASIC的硬件加速器以及低功耗嵌入式GPGPU来评估我们的技术,并表明我们的方法在最终的网络准确性中只会导致很小或可以忽略不计的损失。我们使用MNIST,CIFAR-10和SVHN数据集的三个著名网络分析了我们提出的方法的性能,我们表明,我们能够在三个基准测试中实现高达95%的能量降低,准确性损失小于1%。此外,相比以前的工作动态网络重新配置,我们表明,我们的方法导致约50%的存储需求节省,同时实现类似的准确性。
英文摘要:
We present a novel dynamic configuration technique for deep neural networks that permits step-wise energy-accuracy trade-offs during runtime. Our configuration technique adjusts the number of channels in the network dynamically depending on response time, power, and accuracy targets. To enable this dynamic configuration technique, we co-design a new training algorithm, where the network is incrementally trained such that the weights in channels trained in earlier steps are fixed. Our technique provides the flexibility of multiple networks while storing and utilizing one set of weights. We evaluate our techniques using both an ASIC-based hardware accelerator as well as a low-power embedded GPGPU and show that our approach leads to only a small or negligible loss in the final network accuracy. We analyze the performance of our proposed methodology using three well-known networks for MNIST, CIFAR-10, and SVHN datasets, and we show that we are able to achieve up to 95% energy reduction with less than 1% accuracy loss across the three benchmarks. In addition, compared to prior work on dynamic network reconfiguration, we show that our approach leads to approximately 50% savings in storage requirements, while achieving similar accuracy.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。