Deploying Convolutional Neural Network (CNN) models on ubiquitous Internet of Things (IoT) devices in a cloud-assisted manner to provide users with a variety of high-quality services has become mainstream. Most existing studies speed up model cloud training/on-device inference by reducing the number of convolution (Conv) parameters and floating-point operations (FLOPs). However, they usually employ two or more lightweight operations (e.g., depthwise Conv,$1\times 1$cheap Conv) to replace a Conv, which can still affect the model's speedup even with fewer parameters and FLOPs. To this end, we propose the Grouped NonLinear transformation generation method (GroupNL), leveraging data-agnostic, hyperparameters-fixed, and lightweight Nonlinear Transformation Functions (NLFs) to generate diversified feature maps on demand via grouping, thereby reducing resource consumption while improving the robustness of CNNs. First, in a GroupNL Conv layer, a small set of feature maps, i.e., seed feature maps, are generated based on the seed Conv operation. Then, we split seed feature maps into several groups, each with a set of different NLFs, to generate the required number of diversified feature maps with tensor manipulation operators and nonlinear processing in a lightweight manner without additional Conv operations. We further introduce a sparse GroupNL Conv to speed up by reasonably designing the seed Conv groups between the number of input channels and seed feature maps. Experiments conducted on benchmarks and on-device resource measurements demonstrate that the GroupNL Conv is an impressive alternative to Conv layers in baseline models. Specifically, on Icons-50 dataset, the accuracy of GroupNL-ResNet-18 is 2.86% higher than ResNet-18; on ImageNet-C dataset, the accuracy of GroupNL-EfficientNet-ES achieves about 1.1% higher than EfficientNet-ES. In addition, we verified the efficiency of GroupNL-based models in terms of cloud training and on-device inference.

Deploying quantized deep neural network (DNN) models with resource adaptation capabilities on ubiquitous Internet of Things (IoT) devices to provide high-quality AI services can leverage the benefits of compression and meet multi-scenario resource requirements. However, existing dynamic/mixed precision quantization requires retraining or special hardware, whereas post-training quantization (PTQ) has two limitations for resource adaptation: (i) The state-of-the-art PTQ methods only provide one fixed bitwidth model, which makes it challenging to adapt to the dynamic resources of IoT devices; (ii) Deploying multiple PTQ models with diverse bitwidths consumes large storage resources and switching overheads. To this end, this paper introduces a resource-friendly post-training integer-nesting quantization, i.e., NestQuant, for on-device quantized model switching on IoT devices. The proposed NestQuant incorporates the integer weight decomposition, which bit-wise splits quantized weights into higher-bit and lower-bit weights of integer data types. It also contains a decomposed weights nesting mechanism to optimize the higher-bit weights by adaptive rounding and nest them into the original quantized weights. In deployment, we can send and store only one NestQuant model and switch between the full-bit/part-bit model by paging in/out lower-bit weights to adapt to resource changes and reduce consumption. Experimental results on the ImageNet-1K pretrained DNNs demonstrated that the NestQuant model can achieve high performance in top-1 accuracy, and reduce in terms of data transmission, storage consumption, and switching overheads. In particular, the ResNet-101 with INT8 nesting INT6 can achieve 78.1% and 77.9% accuracy for full-bit and part-bit models, respectively, and reduce switching overheads by approximately 78.1% compared with diverse bitwidths PTQ models. Code:https://github.com/jianhayes/NESTQUANT.

Cloud-assisted Internet of Things (IoT) device deployment of deep neural networks (DNNs) promotes On-device deep learning to provide users with ubiquitous high-quality services by solving the contradiction between insufficient IoT device resources and intensive demand for high-performance DNN resources. However, most existing methods optimize DNNs by considering one or two terms of transmission, computation, and storage resources, but do not consider all three terms at the same time in cloud-assisted IoT device deployment and updating DNNs. To this end, we propose a non-learnable module-based ResNet and a cloud-assisted on-device deep learning framework, ReFrame, based on the consideration of three indicators: model transmission parameters, computation resources, and storage resources. In the proposed method, we first specify that some parameters in DNNs are non-learnable and randomly initialized, so that, these parameters can be saved and reproduced with a few random seeds. By doing so, the cloud only transmits random seeds and learnable parameters to reduce the number of parameter transmissions. Second, we reduce the computation resource consumption of the model by introducing computation-friendly operators, such as pooling, to replace vanilla convolutions. Finally, since random seeds are used to save non-learnable model parameters, on IoT devices we only need to store random seeds and learnable parameters to reproduce the well-trained model. Compared with saving the complete model, our method greatly reduces IoT device storage resource consumption. Experimental results on image classification, object detection, and semantic segmentation tasks demonstrate the effectiveness of the proposed method. Specifically, on the CIFAR-10, our proposed method reduces approximately 89% of FLOPs and 90% of transmitted data in the prototype system compared to ResNet-18.

Deploying a single deep neural network (DNN) model on ubiquitous Internet of Things (IoT) devices to provide multiple vision services (e.g., semantic segmentation service, facial attribute recognition service) has attracted significant interest from industry and academia. However, the most relevant studies have two limitations: (i) the multiple services DNN commonly has low performance due to the gradient interference between different vision services; (ii) the multiple services DNN has low inference speed on IoT devices. To this end, this paper introduces a multiple vision services acceleration framework for IoT devices. The proposed MCBNet contains a novel multi-controllable branching structure to control different services explicitly for improving multiple services performance and a service balance layer with the per-channel trainable masks in specific branches for mitigating gradient interference. In deployment, the proposed framework converts the multi-branch topology MCBNet model into the plain topology Rep-MCBNet model by re-parameterization technique for accelerating services inference speed. Experimental results on the CelebA and Cityscapes datasets show that our proposed framework outperforms the state-of-the-art approaches in terms of performance and inference speed. In particular, we deploy the Rep-MCBNet on the laptop with Intel mobile processors demonstrating that the proposed framework achieves 1.9× faster processing frames per second than ETR-NLP, with 1.5× fewer parameters and 1.3× fewer floating-point operations.