ENOS: Energy-Aware Network Operator Search in Deep Neural Networks
Shamma Nasrin, Ahish Shylendra, Nastaran Darabi, Theja Tulabandhula, Wilfred Gomes, Ankush Chakrabarty, Amit Ranjan Trivedi
Abstract
This work proposes a novel <i>Energy-aware Network Operator Search</i> (<i>ENOS</i>) approach to address the energy-accuracy trade-offs of a deep neural network (DNN) accelerator. In recent years, novel hardware-friendly inference operators such as binary-weight, multiplication-free, and deep-shift have been proposed to improve the computational efficiency of a DNN accelerator. However, simplifying DNN operators invariably comes at lower accuracy, especially on complex processing tasks. While prior works generally implement the same inference operator throughout the neural architecture, the proposed ENOS framework allows an <i>optimal layer-wise integration of inference operators with optimal precision</i> to maintain high prediction accuracy and high energy efficiency. The search in ENOS is formulated as a continuous optimization problem, solvable using gradient descent methods, thereby minimally increasing the training cost when learning <i>both</i> layer-wise inference operators and weights. Utilizing ENOS, we discuss multiply-accumulate (MAC) cores for digital spatial architectures that can be reconfigured to different operators and varying computing precision. ENOS training methods with single and bi-level optimization objectives are discussed and compared. We also discuss a sequential operator assignment strategy in ENOS that only learns the assignment for one layer in one training step. Furthermore, a stochastic mode of ENOS is also presented. ENOS is characterized on ShuffleNet and SqueezeNet using CIFAR10 and CIFAR100. Compared to the conventional uni-operator approaches, under the same energy budget, ENOS improves accuracy by 10–20%. ENOS also outperforms the accuracy of comparable mixed-precision uni-operator implementations by 3-5% for the same energy budget.