This research investigates the integration of Metaheuristic Algorithms (MAs) with the Extreme Learning Machine (ELM) model to optimize parameters of activationfunction. While MAs have traditionally been employed for ...
详细信息
ISBN:
(纸本)9789492859280
This research investigates the integration of Metaheuristic Algorithms (MAs) with the Extreme Learning Machine (ELM) model to optimize parameters of activationfunction. While MAs have traditionally been employed for weights selection, a methodology that utilizes MA for the selection of activationfunction parameters was proposed. The performance of 24 distinctive activationfunctions was evaluated on diverse and widespread benchmark datasets: Ionosphere, Breast Cancer, Australian Credits, Musk and Banana. The results demonstrate a strong dependence on selecting an optimal activationfunction for each task, with variations in accuracy ranging up to 60 percentage points. The MA-ELM approach shows promising results, providing improved accuracy and reducing the number of required neurons in certain cases. The approach offers an efficient alternative to the typical MA-ELM method, requiring evaluation of only a few parameter values compared to the optimization of hundreds or thousands of weights. This approach enhances the generalization abilities of core ELM method and reduces computational time in comparison to typical MA-ELM. These findings validate the effectiveness of the proposed MA-ELM approach, contributing to the understanding of integrating MA with activationfunctions in ELM and offering insights for enhancing model performance in various applications.
The recognition of personnel entering and leaving the mine is an important link to ensure safe production. As an effective identity recognition technology, face recognition has been widely and deeply studied while fac...
详细信息
ISBN:
(纸本)9789811979422;9789811979439
The recognition of personnel entering and leaving the mine is an important link to ensure safe production. As an effective identity recognition technology, face recognition has been widely and deeply studied while facing the problem that the recognition rate is not high in the complex and harsh environment of mine such as facial expressions, pose variation, and low-resolution of face images. For effectively improving the face recognition rate of miners in uneven illumination environment, a face recognition method based on inverted residual network is proposed. In this method, through the optimization of the activationfunction, the amount of calculation can be greatly reduced while keeping the almost equivalent performance. And by the fusion of inverted residual network, the problem of partial feature information loss in face image recognition model training is effectively solved, which greatly improves the accuracy of recognition. The experimental results show that the accuracy of the inverted residual face recognition model is 81.4%, which is 5.7% higher than the residual network algorithm with additional 4.3% of time overhead, and 9.9% higher than the MTCNN model with only the 1/13 recognition time of MTCNN.
Long Short -Term Memory Networks (LSTMs) are pivotal in on -device time series analysis for embedded systems, particularly for managing sensor data streams. Yet, their deployment on resource -constrained embedded devi...
详细信息
Long Short -Term Memory Networks (LSTMs) are pivotal in on -device time series analysis for embedded systems, particularly for managing sensor data streams. Yet, their deployment on resource -constrained embedded devices presents notable challenges. In response, we introduce a novel parameterized architecture for LSTM accelerators designed explicitly for embedded Field -Programmable Gate Arrays (FPGAs). Our approach involves strategic design choices, such as employing computationally efficient activationfunctions and optimizing clock frequency with a pipelined Arithmetic Logic Unit (ALU). These decisions drive our architecture towards enhanced energy efficiency while maintaining adaptability across diverse application scenarios. A key feature of our architecture is its configurable parameters, which allow for tailored optimization through the optional use of Digital Signal Processor Slices for ALUs and the selective implementation of activationfunctions. Our empirical evaluations conducted on the Spartan -7 XC7S15 FPGA demonstrate the robustness of our methodology, achieving a 2.33 x improvement in energy efficiency over previous solutions. Furthermore, our study examines the correlation between memory resource types and energy efficiency across various LSTM model sizes. Impressively, even with a 9 x increase in the hidden size of the LSTM cell, our accelerator maintains an energy efficiency of 10.03 GOP/s/W, with only a minor decrease of 14.65%. However, it is critical to note that our current design is not yet optimized for larger FPGA models such as the Spartan -7 XC7S25 and XC7S50 . For these models, timing constraints, rather than resource limitations, pose challenges to scaling, highlighting a potential area for future optimization.
暂无评论