DNA-TEQ: An Adaptive Exponential Quantization of Tensors for DNN Inference
This work addresses the need for efficient DNN deployment on embedded systems by enabling lower bit-width quantization without retraining, though it is incremental as it builds on existing quantization techniques.
The paper tackles the problem of high performance loss when reducing numerical precision below 8 bits in DNN inference by proposing DNA-TEQ, an adaptive exponential quantization method that achieves an average 40% compression ratio over INT8 baseline with negligible accuracy loss and 66% energy savings.
Quantization is commonly used in Deep Neural Networks (DNNs) to reduce the storage and computational complexity by decreasing the arithmetical precision of activations and weights, a.k.a. tensors. Efficient hardware architectures employ linear quantization to enable the deployment of recent DNNs onto embedded systems and mobile devices. However, linear uniform quantization cannot usually reduce the numerical precision to less than 8 bits without sacrificing high performance in terms of model accuracy. The performance loss is due to the fact that tensors do not follow uniform distributions. In this paper, we show that a significant amount of tensors fit into an exponential distribution. Then, we propose DNA-TEQ to exponentially quantize DNN tensors with an adaptive scheme that achieves the best trade-off between numerical precision and accuracy loss. The experimental results show that DNA-TEQ provides a much lower quantization bit-width compared to previous proposals, resulting in an average compression ratio of 40% over the linear INT8 baseline, with negligible accuracy loss and without retraining the DNNs. Besides, DNA-TEQ leads the way in performing dot-product operations in the exponential domain, which saves 66% of energy consumption on average for a set of widely used DNNs.