You might also read
Articles linked to this work by shared authors, journal, and citation graph.
This study introduces dynamic learning rate optimization for backpropagation (BP) algorithms, significantly accelerating convergence. The new methods reduce running time by 10-50 times without increasing computational load.
Area of Science:
Background:
Purpose of the Study:
Main Methods:
Main Results:
Conclusions: