Related Concept Videos
Radial System Protection
In a radial system with a fault downstream of the third breaker, ideally, only the third breaker will open, isolating the fault and interrupting the load connected beyond it. The second breaker has a longer delay setting,...
Radius of Gyration of an Area
Rectangular and Triangular Pulse Function
For example, consider a rectangular pulse with a 5V amplitude, a 3-second duration, and centered at t=2 seconds. This pulse can be expressed using the rectangular function, written as,
Radical Chain-Growth Polymerization: Overview
Cylinders in Three-Dimensional Space
Parallel Processing
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
Distinct miRNA profiles in human amniotic tissue and its vesicular and non-vesicular secretome.
Successful Trial of Labor After Two Caesarean Sections (TOLA2C): Analysis of a Delivery Protocol with Feto-Maternal Outcome.
A new paradigm for estimating the prevalence of plastic litter in the marine environment.
Related Experiment Videos
Radial basis function networks GPU-based implementation.
Andreas Brandstetter1, Alessandro Artusi
1Institute of Computer Graphics and Algorithms, Vienna University of Technology, Vienna A1040, Austria.
This study presents a graphic processing unit (GPU) implementation for accelerating radial basis function network (RBFN) training. The GPU approach significantly reduces computational costs compared to central processing unit (CPU) methods.
Area of Science:
- Artificial Intelligence
- Computational Science
- Hardware Acceleration
Background:
- Neural networks (NNs) offer great potential but face limitations, notably long training times.
- Traditional hardware implementations (CPU) for NNs are costly and inflexible.
- Graphic Processing Units (GPUs) have emerged as powerful, versatile computing resources.
Purpose of the Study:
- To accelerate the training process of Radial Basis Function Networks (RBFNs).
- To overcome the time performance limitations of RBFNs in dynamic application domains.
- To demonstrate the efficacy of GPU implementation for RBFN learning.
Main Methods:
- Developed a complete graphic processing unit (GPU) implementation for the RBFN learning process.
- Compared the computational cost and training time against traditional central processing unit (CPU) implementations.
- Focused on optimizing the entire learning procedure for parallel processing on GPUs.
Main Results:
- Achieved a reduction in computational cost by approximately two orders of magnitude.
- Demonstrated significant acceleration of the RBFN training process using GPU acceleration.
- Validated the feasibility and efficiency of GPU-based RBFN learning.
Conclusions:
- GPU implementation drastically reduces RBFN training time and computational cost.
- This approach enhances the applicability of RBFNs in time-sensitive domains.
- Accelerated RBFN learning on GPUs offers a practical solution for complex computational tasks.