In order to solve the large-scale, strong coupling, and nonlinear optimization problems in many Internet-of-Things (IoT) applications, such as intelligent infrastructure and smart city, this article proposes a real-coded genetic algorithm based on an auxiliary-individual-directed crossover operator (AIDX). AIDX is an alternative offspring framework of the directed crossover which uses auxiliary individual to reduce the search space of alternative offspring for the rapid optimization of multidimensional problems. In our solution AIDX, the parents-center distribution is adopted to reduce the risk of convergence to local optimization and enhance the stability of the algorithm. In addition, in order to increase the diversity of the population at the late stage of optimization, K-Bit-Swap (KBS) is used as a supplement for the exchange of genetic information among individuals in different dimensions. In the extensive experiments, 24 benchmarks with different dimensions are used to evaluate the performance of AIDX-GA. The results show that the proposed AIDX-GA has a significantly improved optimization effect on multidimensional optimization problems, and the stability of the algorithm is largely enhanced even when the global optimum is located near the boundary. AIDX-GA has also been evaluated in an industrial IoT case that identifies the resistance coefficients of a pipe network in the city infrastructure. The results show that the accuracy of AIDX-GA is high, and it has excellent universality and stability, which can be used to solve many IoT problems.

While Deep Neural Networks (DNNs) are becoming increasingly popular, there is a growing trend to accelerate the DNN applications on hardware platforms like GPUs, FPGAs, etc., to gain higher performance and efficiency. However, it is time-consuming to tune the performance for such platforms due to the large design space and the expensive cost to evaluate each design point. Although many tuning algorithms, such as XGBoost tuner and genetic algorithm (GA) tuner, have been proposed to guide the design space exploring process in the previous work, the timing issue still remains a critical problem. In this work, we propose a novel auto-tuning framework to optimize the DNN operator design on GPU by leveraging the tuning history efficiently in different scenarios. Our experiments show that we can achieve superior performance than the state-of-the-art work, such as auto-tuning framework TVM and the handcraft optimized library cuDNN, while reducing the searching time by 8.96x and 4.58x comparing with XGBoost tuner and GA tuner in TVM.