Search

Quantized neural networks (QNNs) have become a standard operation for efficiently deploying deep learning models on hardware platforms in real application scenarios. An empirical study on German traffic sign recognition benchmark (GTSRB) dataset shows that under the three white-box adversarial attacks of fast gradient sign method, random + fast gradient sign method and basic iterative method, the accuracy of the full quantization model was only 55%, much lower than that of the full precision model (73%). This indicates the adversarial robustness of the full quantization model is much worse than that of the full precision model. To improve the adversarial robustness of the full quantization model, we have designed an adversarial attack defense platform based on field-programmable gate array (FPGA) to jointly optimize the efficiency and robustness of QNNs. Various hardware-friendly techniques such as adversarial training and feature squeezing were studied and transferred to the FPGA platform based on the designed accelerator of QNN. Experiments on the GTSRB dataset show that the adversarial training embedded on FPGA can increase the model's average accuracy by 2.5% on clean data, 15% under white-box attacks, and 4% under black-box attacks, respectively, demonstrating our methodology can improve the robustness of the full quantization model under different adversarial attacks.

Recent advancements in EEG-based BCI technologies have been explored to assist individuals in executing brain-to-robot tasks. In the future, using BCI systems in both personal and professional domains is anticipated in widespread adoption. One promising application is facing home life, to enable people to use commercial EEG equipment to implement daily tasks. However, current BCI studies mainly focus on single-brain-to-single-robot interaction, which has limitations in representing diverse human intentions. A generalized BCI system with multiple EEG devices allows users to more precisely or collaboratively control robots. Therefore, it is imperative to extend the BCI techniques to future collaboration scenarios. In this paper, we present a new system, NeuroBCI, for multi-brain-to-multi-robot interaction through the integration of sensing, computing, communication, and control. To improve sensing efficiency, NeuroBCI employs a sparse attention mechanism to extract joint features from heterogeneous EEG data. Parallel computation and transmission for multi-user multi-task scenarios are handled by semantic autoencoder and autodecoder communications. A code map is designed to ensure concurrent control and model compression methods are used on both transmitter and receiver sides. Our experiments in comparison with state-of-the-art works show the superior performance of NeuroBCI on sensing, computing, communication, and control as a holistic system.

With the development of artificial intelligence and the improvement of hardware computing power, deep learning models have become widely used in the Internet of Things (IoT) field, especially for analyzing spatiotemporal data collected by wireless sensors. Recurrent neural networks (RNNs) such as long short-term memory (LSTM) network are generally used for these time-series data. Hyperparameter settings of model training are regarded as essential factors for the performance of deep learning models. Manually optimizing hyperparameters not only cost more resources but also be more likely to set hyperparameters that follow stereotypes, resulting in unreasonable hyperparameter settings and poor model performance. As one of the most important fields in automated machine learning research, automated hyperparameter optimization (HPO) mainly includes grid search, hyperparameter search based on genetic algorithm, etc. Whereas these methods have their own drawbacks. In this article, an automated HPO method based on parallel genetic algorithm (PGA) is proposed. According to the process of PGA, this article divided HPO into several stages, including population initialization, fitness function, tournament selection, crossover operators, mutation operators, subgroup exchange, and end of evolution. Then, the proposed HPO method is implemented in LSTM models and tested on two different time-series datasets collected by real-world IoT sensors. By comparing our proposed method with other mainstream HPO methods in different datasets, it is proved that our HPO method based on PGA shows a better performance on both time costs and prediction results.