View source for US Patent Application 18246782. LEARNING MODEL OPTIMIZATION DEVICE, LEARNING MODEL OPTIMIZATION METHOD, AND LEARNING MODEL OPTIMIZATION PROGRAM simplified abstract

=LEARNING MODEL OPTIMIZATION DEVICE, LEARNING MODEL OPTIMIZATION METHOD, AND LEARNING MODEL OPTIMIZATION PROGRAM=


==Organization Name==

[[Nippon Telegraph and Telephone Corporation]]


==Inventor(s)==

[[Daiki Chijiwa of Musashino-shi, Tokyo (JP)]]

[[Kenji Umakoshi of Musashino-shi, Tokyo (JP)]]

[[Tomohiro Inoue of Musashino-shi, Tokyo (JP)]]

[[Daigoro Yokozeki of Musashino-shi, Tokyo (JP)]]

==LEARNING MODEL OPTIMIZATION DEVICE, LEARNING MODEL OPTIMIZATION METHOD, AND LEARNING MODEL OPTIMIZATION PROGRAM - A simplified explanation of the abstract==


This abstract first appeared for US patent application 18246782 titled 'LEARNING MODEL OPTIMIZATION DEVICE, LEARNING MODEL OPTIMIZATION METHOD, AND LEARNING MODEL OPTIMIZATION PROGRAM

==Simplified Explanation==

The patent application describes a learning model optimization device that improves the efficiency of machine learning algorithms by optimizing a binarization matrix.
* The device includes a binarization matrix setting unit that sets a binarization matrix with elements as either "0" or "1".
* A transformed matrix setting unit sets a transformed matrix by multiplying each element of the parameter matrix with the corresponding element of the binarization matrix.
* A learning unit performs machine learning using the transformed matrix and adjusts the numerical values of the binarization matrix to optimize it.
* A re-randomization processing unit further adjusts the parameter matrix to improve the optimization process.


==Original Abstract Submitted==

A learning model optimization device includes a binarization matrix setting unit configured to set a binarization matrix m in which each element is a numerical value of “0” or “1” and a transformed matrix setting unit configured to set a transformed matrix M having, as an element, a product of each element of the parameter matrix and each element of the binarization matrix in the same row and the same column. The learning model optimization device further includes a learning unit configured to perform machine learning using the transformed matrix M and change a numerical value of each element of the binarization matrix m such that a result of the machine learning approaches teacher data, thereby optimizing the binarization matrix m, and a re-randomization processing unit configured to change again a parameter of the parameter matrix w.