Search

Binarization is an extreme quantization technique that is attracting research in the Internet of Things (IoT) field, as it radically reduces the memory footprint of deep neural networks without a correspondingly significant accuracy drop. To support the effective deployment of Binarized Neural Networks (BNNs), we propose CBin-NN, a library of layer operators that allows the building of simple yet flexible convolutional neural networks (CNNs) with binary weights and activations. CBin-NN is platform-independent and is thus portable to virtually any software-programmable device. Experimental analysis on the CIFAR-10 dataset shows that our library, compared to a set of state-of-the-art inference engines, speeds up inference by 3.6 times and reduces the memory required to store model weights and activations by 7.5 times and 28 times, respectively, at the cost of slightly lower accuracy (2.5%). An ablation study stresses the importance of a Quantized Input Quantized Kernel Convolution layer to improve accuracy and reduce latency at the cost of a slight increase in model size. [ABSTRACT FROM AUTHOR]

Internet of Things technologies are spurring new types of instructional games, namely reality-enhanced serious games (RESGs), that support training directly in the field. This paper investigates a key feature of RESGs, i.e., user performance evaluation using real data, and studies an application of RESGs for promoting fuel-efficient driving, using fuel consumption as an indicator of driver performance. In particular, we propose a reference model for supporting a novel smart sensing dataflow involving the combination of two modules, based on machine learning, to be employed in RESGs in parallel and in real-time. The first module concerns quantitative performance assessment, while the second one targets verbal recommendation. For the assessment module, we compared the performance of three well-established machine learning algorithms: support vector regression, random forest and artificial neural networks. The experiments show that random forest achieves a slightly better performance assessment correlation than the others but requires a higher inference time. The instant recommendation module, implemented using fuzzy logic, triggers advice when inefficient driving patterns are detected. The dataflow has been tested with data from the enviroCar public dataset, exploiting on board diagnostic II (OBD II) standard vehicular interface information. The data covers various driving environments and vehicle models, which makes the system robust for real-world conditions. The results show the feasibility and effectiveness of the proposed approach, attaining a high estimation correlation (R2 = 0.99, with random forest) and punctual verbal feedback to the driver. An important word of caution concerns users' privacy, as the modules rely on sensitive personal data, and provide information that by no means should be misused. [ABSTRACT FROM AUTHOR]

This paper presents the Edge Learning Machine (ELM), a machine learning framework for edge devices, which manages the training phase on a desktop computer and performs inferences on microcontrollers. The framework implements, in a platform-independent C language, three supervised machine learning algorithms (Support Vector Machine (SVM) with a linear kernel, k-Nearest Neighbors (K-NN), and Decision Tree (DT)), and exploits STM X-Cube-AI to implement Artificial Neural Networks (ANNs) on STM32 Nucleo boards. We investigated the performance of these algorithms on six embedded boards and six datasets (four classifications and two regression). Our analysis—which aims to plug a gap in the literature—shows that the target platforms allow us to achieve the same performance score as a desktop machine, with a similar time latency. ANN performs better than the other algorithms in most cases, with no difference among the target devices. We observed that increasing the depth of an NN improves performance, up to a saturation level. k-NN performs similarly to ANN and, in one case, even better, but requires all the training sets to be kept in the inference phase, posing a significant memory demand, which can be afforded only by high-end edge devices. DT performance has a larger variance across datasets. In general, several factors impact performance in different ways across datasets. This highlights the importance of a framework like ELM, which is able to train and compare different algorithms. To support the developer community, ELM is released on an open-source basis. [ABSTRACT FROM AUTHOR]