Real-time fast learning hardware implementation

نویسندگان

چکیده

Machine learning algorithms are widely used in many intelligent applications and cloud services. Currently, the hottest topic this field is Deep Learning represented often by neural network structures. fully known as deep network, artificial a typical machine method an important way of learning. With massive growth data, research has made significant achievements natural language processing (NLP), image recognition, autonomous driving. However, there still breakthroughs needed training time energy consumption Based on our previous fast architecture for paper, solution to minimize connected analysed theoretically. Therefore, we propose new parallel algorithm structure with over-tuned parameters. This strategy finally leads adaptation delay impact performance analyzed using simple benchmark case study. It shown that reduction step size could be proposed compensate errors due delayed adaptation, then gain phase function parameters chosen Finally, realize real-time learning, implemented FPGA parallelism flexibility, integration shows good low power consumption.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Real-time Median Filtering with a Fast Hardware Sorter

This work develops a hardware realization of a median filter suitable for use in real-time control and monitoring applications in power electronic circuits. Median filters have the ability to supprem impulse noise in signals while preserving underlying edges. The performance of the filter is demonstrated with results from a prototype.

متن کامل

A real-time shot cut detector: Hardware implementation

With the enormous growth in digital audiovisual (AV) information in our life, there is an important need for tools which enable describing the AV content information. In this context, the MPEG-7 standard was developed in order to provide a set of standardized description tools which generate metadata about AV content. However, before any content-based manipulations, the hierarchical structure o...

متن کامل

Technical Report: Real-Time Aware Hardware Implementation Effort Estimation

This paper presents a structured method and underlying models for estimating the hardware implementation effort of hard real-time constrained embedded systems. We propose an optimization model which takes some of the most common optimization techniques into account as well as the order in which they should be applied. We suggest a set of two metrics used to characterise the effects of optimisat...

متن کامل

Hardware Implementation of a Real Time Image Compression

Real time processing of image deals with applying all required operations within a range of time not exceed the acceptable time of human eyes. Real time processing does not realized on standalone computer, so it need special hardware. The big challenge of image processing work is the processing time, in which there are big amount of data (pixels) that must be processed at a specific time. The m...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: International Journal for Simulation and Multidisciplinary Design Optimization

سال: 2023

ISSN: ['1779-627X', '1779-6288']

DOI: https://doi.org/10.1051/smdo/2023001