学習システム、学習装置、および学習方法

Japanese Patent JP7468540

B2

A learning system includes teacher DNN feature extraction unit extracting a feature of each of a plurality of training data, teacher DNN estimate calculation unit calculating a first estimate of a label corresponding to each of the training data, student DNN feature extraction unit extracting a feature of each of the training data, student DNN estimate calculation unit calculating a second estimate of a label corresponding to each of the training data, noisy label correction unit determining whether or not the label corresponding to the training data is a label including a noise, based on the label corresponding to the training data and the first estimate, and update unit updating weights in the student DNN so as to reduce a difference between the feature extracted by the teacher DNN feature extraction unit and the feature extracted by the student DNN feature extraction unit while decreasing an influence of the label including the noise.

WO/2021/231061	DEVICE DEACTIVATION BASED ON BEHAVIOR PATTERNS
WO/2018/013982	CLASSIFYING IMAGES USING MACHINE LEARNING MODELS
WO/2021/197357	PROJECTION FOCUSING METHOD AND PROJECTION FOCUSING DEVICE

Ryo Takamoto

JP2021550747A

April 16, 2024

September 30, 2019

Click for automatic bibliography generation  

NEC

G06N3/08

WANG, Zihan ほか,CrossWeigh: Training Named Entity Tagger from Imperfect Annotations,arXiv[online],2019年09月03日,[retrieved on 2023.10.30], Retrieved from the Internet:
ROMERO, Adriana ほか,FitNets: Hints for Thin Deep Nets,arXiv[online],2015年03月27日,pp.1-13,[retrieved on 2019.12.10], Retrieved from the Internet:
ラベルの付け間違いが分かる!?,The Labellio Blog[online],2019年07月04日,[retrieved on 2019.12.10], Retrieved from the Internet: yuzupepper,論文メモ:Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approach,Qiita[online],2017年08月20日,[retrieved on 2019.12.10], Retrieved from the Internet:

Rock wall Fuyuki
Masato Shiokawa

Previous Patent: stator and motor

Next Patent: dialysis machine