no code implementations • 25 Oct 2021 • Shin Kamada, Takumi Ichimura
In our research, an adaptive structural learning method of Restricted Boltzmann Machine (Adaptive RBM) and Deep Belief Network (Adaptive DBN) have been developed as a deep learning model.
no code implementations • 25 Oct 2021 • Shin Kamada, Takumi Ichimura
Our Adaptive DBN had an advantage of not only the detection accuracy but also the inference time compared with the conventional CNN in the experiment results.
no code implementations • 25 Oct 2021 • Takumi Ichimura, Shin Kamada
In order to represent such cases, this paper investigated a distillation learning model of Adaptive DBN.
no code implementations • 25 Oct 2021 • Shin Kamada, Takumi Ichimura, Takashi Iwasaki
The dataset contains about 56, 000 crack images for three types of concrete structures: bridge decks, walls, and paved roads.
no code implementations • 30 Sep 2019 • Takumi Ichimura, Shin Kamada
In order to distinguish such cases, this paper investigated a re-learning model of Adaptive DBN with two or more child models, where the original trained model can be seen as a parent model and then new child models are generated for some misclassified cases.
no code implementations • 30 Sep 2019 • Shin Kamada, Takumi Ichimura
In this paper, a new object detection method for the DBN architecture is proposed for localization and category of objects.
no code implementations • 30 Sep 2019 • Shin Kamada, Takumi Ichimura
The method can find the optimal number of hidden neurons of a Restricted Boltzmann Machine (RBM) by neuron generation-annihilation algorithm to train the given input data, and then it can make a new layer in DBN by the layer generation algorithm to actualize a deep data representation.
no code implementations • 27 Aug 2018 • Shin Kamada, Takumi Ichimura, Toshihide Harada
The adaptive structural learning method of Deep Belief Network (DBN) has been developed.
no code implementations • 11 Jul 2018 • Shin Kamada, Takumi Ichimura
We can success the knowledge extraction from the trained deep learning with high classification capability.
no code implementations • 11 Jul 2018 • Takumi Ichimura, Shin Kamada
Deep Learning has the hierarchical network architecture to represent the complicated features of input patterns.
no code implementations • 11 Jul 2018 • Shin Kamada, Takumi Ichimura
For this reason, not only the numerical data and text data but also the time-series data are transformed to the image data format.
no code implementations • 10 Jul 2018 • Shin Kamada, Takumi Ichimura
In this paper, we propose the adaptive learning method of DBN that can determine the optimal number of layers during the learning.
no code implementations • 10 Jul 2018 • Shin Kamada, Takumi Ichimura
In this method, a new hidden neuron is generated if the energy function is not still converged and the variance of the parameters is large.
no code implementations • 10 Jul 2018 • Shin Kamada, Takumi Ichimura
As a result, the classification capability can achieve a great success (97. 1\% to unknown data set).
no code implementations • 9 Apr 2018 • Shin Kamada, Takumi Ichimura
However, some landmarks was not detected correctly by the previous method because they didn't have enough amount of information for the feature extraction.
no code implementations • 8 Apr 2018 • Takumi Ichimura, Shin Kamada
The clonal selection principle explains the basic features of an adaptive immune response to a antigenic stimulus.
no code implementations • 8 Apr 2018 • Takumi Ichimura, Shin Kamada
Mobile Phone based Participatory Sensing (MPPS) system involves a community of users sending personal information and participating in autonomous sensing through their mobile phones.