18527143. ADAPTATION OF NEURAL NETWORKS TO NEW OPERATING SITUATIONS simplified abstract (Robert Bosch GmbH)
Contents
ADAPTATION OF NEURAL NETWORKS TO NEW OPERATING SITUATIONS
Organization Name
Inventor(s)
Jan Hendrik Metzen of Boeblingen (DE)
ADAPTATION OF NEURAL NETWORKS TO NEW OPERATING SITUATIONS - A simplified explanation of the abstract
This abstract first appeared for US patent application 18527143 titled 'ADAPTATION OF NEURAL NETWORKS TO NEW OPERATING SITUATIONS
The patent application describes a method for adapting a neural network to process new measurement data after being trained on existing examples.
- A working space is provided for low-dimensional representations of parameter vectors.
- An image assigns each low-dimensional representation a parameter vector.
- Candidate representations are set up in the working space.
- Candidate representations are translated into candidate parameter vectors using the image.
- A predetermined quality function is evaluated for each candidate parameter vector based on the neural network's output for the new data.
- The candidate parameter vector with the best quality function value is considered the optimal adaptation for the neural network.
Potential Applications: - Industrial automation - Medical diagnostics - Financial forecasting
Problems Solved: - Efficient adaptation of neural networks to new data - Improved accuracy in processing measurement data
Benefits: - Enhanced performance of neural networks - Faster adaptation to changing data sets - Increased reliability in data processing
Commercial Applications: Title: "Advanced Neural Network Adaptation Technology for Enhanced Data Processing" This technology can be utilized in various industries such as manufacturing, healthcare, and finance to improve data analysis and decision-making processes.
Prior Art: Research on neural network adaptation methods and quality functions in data processing.
Frequently Updated Research: Stay updated on advancements in neural network adaptation techniques and quality function optimization for data processing.
Questions about Neural Network Adaptation Technology: 1. How does this method improve the efficiency of neural network adaptation? 2. What are the key factors influencing the evaluation of candidate parameter vectors in this process?
Original Abstract Submitted
A method for adapting a neural network for processing measurement data, which network has been trained on training examples, and whose behavior is characterized by a parameter vector, to a new record of measurement data. In the method: a working space for low-dimensional representations of parameter vectors and an image which assigns to each low-dimensional representation a parameter vector are provided; in the working space, candidate representations are set up; the candidate representations are translated using the image into candidate parameter vectors; for each candidate parameter vector, a predetermined quality function is evaluated, which depends on the output of the neural network for the record in the state in which the candidate parameter vector has replaced the original parameter vector; a candidate parameter vector, for which the quality function assumes the best value, is evaluated as the optimal adaptation of the parameters of the neural network to the record.
