Lightweight Compression of Intermediate Neural Network Features for Collaborative Intelligence
This addresses the need for efficient edge-cloud split DNNs in applications like mobile or edge devices, offering a lightweight compression method without retraining, though it is incremental as it builds on existing quantization techniques.
The paper tackles the problem of compressing intermediate neural network features for collaborative intelligence, achieving compression of 32-bit floating point activations down to 0.6 to 0.8 bits with less than 1% accuracy loss and outperforming HEVC by up to 1.3% in inference accuracy.
In collaborative intelligence applications, part of a deep neural network (DNN) is deployed on a lightweight device such as a mobile phone or edge device, and the remaining portion of the DNN is processed where more computing resources are available, such as in the cloud. This paper presents a novel lightweight compression technique designed specifically to quantize and compress the features output by the intermediate layer of a split DNN, without requiring any retraining of the network weights. Mathematical models for estimating the clipping and quantization error of ReLU and leaky-ReLU activations at this intermediate layer are developed and used to compute optimal clipping ranges for coarse quantization. We also present a modified entropy-constrained design algorithm for quantizing clipped activations. When applied to popular object-detection and classification DNNs, we were able to compress the 32-bit floating point intermediate activations down to 0.6 to 0.8 bits, while keeping the loss in accuracy to less than 1%. When compared to HEVC, we found that the lightweight codec consistently provided better inference accuracy, by up to 1.3%. The performance and simplicity of this lightweight compression technique makes it an attractive option for coding an intermediate layer of a split neural network for edge/cloud applications.