Shin Kamada

Shin Kamada, Takumi Ichimura

An adaptive structural learning method of Restricted Boltzmann Machine (RBM) and Deep Belief Network (DBN) has been developed as one of prominent deep learning models. The neuron generation-annihilation algorithm in RBM and layer generation algorithm in DBN make an optimal network structure for given input during the learning. In this paper, our model is applied to an automatic recognition method of road network system, called RoadTracer. RoadTracer can generate a road map on the ground surface from aerial photograph data. A novel method of RoadTracer using the Teacher-Student based ensemble learning model of Adaptive DBN is proposed, since the road maps contain many complicated features so that a model with high representation power to detect should be required. The experimental results showed the detection accuracy of the proposed model was improved from 40.0\% to 89.0\% on average in the seven major cities among the test dataset. In addition, we challenged to apply our method to the detection of available roads when landslide by natural disaster is occurred, in order to rapidly obtain a way of transportation. For fast inference, a small size of the trained model was implemented on a small embedded edge device as lightweight deep learning. We reported the detection results for the satellite image before and after the rainfall disaster in Japan.

Takumi Ichimura, Shin Kamada

Deep Learning has a hierarchical network architecture to represent the complicated feature of input patterns. We have developed the adaptive structure learning method of Deep Belief Network (DBN) that can discover an optimal number of hidden neurons for given input data in a Restricted Boltzmann Machine (RBM) by neuron generation-annihilation algorithm, and can obtain the appropriate number of hidden layers in DBN. In this paper, our model is applied to a facial expression image data set, AffectNet. The system has higher classification capability than the traditional CNN. However, our model was not able to classify some test cases correctly because human emotions contain many ambiguous features or patterns leading wrong answer by two or more annotators who have different subjective judgment for a facial image. In order to represent such cases, this paper investigated a distillation learning model of Adaptive DBN. The original trained model can be seen as a parent model and some child models are trained for some mis-classified cases. For the difference between the parent model and the child one, KL divergence is monitored and then some appropriate new neurons at the parent model are generated according to KL divergence to improve classification accuracy. In this paper, the classification accuracy was improved from 78.4% to 91.3% by the proposed method.

Shin Kamada, Takumi Ichimura, Takashi Iwasaki

We have developed an adaptive structural Deep Belief Network (Adaptive DBN) that finds an optimal network structure in a self-organizing manner during learning. The Adaptive DBN is the hierarchical architecture where each layer employs Adaptive Restricted Boltzmann Machine (Adaptive RBM). The Adaptive RBM can find the appropriate number of hidden neurons during learning. The proposed method was applied to a concrete image benchmark data set SDNET2018 for crack detection. The dataset contains about 56,000 crack images for three types of concrete structures: bridge decks, walls, and paved roads. The fine-tuning method of the Adaptive DBN can show 99.7%, 99.7%, and 99.4% classification accuracy for three types of structures. However, we found the database included some wrong annotated data which cannot be judged from images by human experts. This paper discusses consideration that purses the major factor for the wrong cases and the removal of the adversarial examples from the dataset.

Shin Kamada, Takumi Ichimura

Deep learning has been a successful model which can effectively represent several features of input space and remarkably improve image recognition performance on the deep architectures. 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. The models have a self-organize function which can discover an optimal number of hidden neurons for given input data in a RBM by neuron generation-annihilation algorithm, and can obtain an appropriate number of RBM as hidden layers in the trained DBN. The proposed method was applied to a concrete image benchmark data set SDNET 2018 for crack detection. The dataset contains about 56,000 crack images for three types of concrete structures: bridge decks, walls, and paved roads. The fine-tuning method of the Adaptive DBN can show 99.7%, 99.7%, and 99.4% classification accuracy for test dataset of three types of structures. In this paper, our developed Adaptive DBN was embedded to a tiny PC with GPU for real-time inference on a drone. For fast inference, the fine tuning algorithm also removed some inactivated hidden neurons to make a small model and then the model was able to improve not only classification accuracy but also inference speed simultaneously. The inference speed and running time of portable battery charger were evaluated on three kinds of Nvidia embedded systems; Jetson Nano, AGX Xavier, and Xavier NX.

Shin Kamada, Takumi Ichimura

In our research, an adaptive structural learning method of Restricted Boltzmann Machine (RBM) and Deep Belief Network (DBN) has been developed as one of prominent deep learning models. The neuron generation-annihilation in RBM and layer generation algorithms in DBN make an optimal network structure for given input during the learning. In this paper, our model is applied to an automatic recognition method of road network system, called RoadTracer. RoadTracer can generate a road map on the ground surface from aerial photograph data. In the iterative search algorithm, a CNN is trained to find network graph connectivities between roads with high detection capability. However, the system takes a long calculation time for not only the training phase but also the inference phase, then it may not realize high accuracy. In order to improve the accuracy and the calculation time, our Adaptive DBN was implemented on the RoadTracer instead of the CNN. The performance of our developed model was evaluated on a satellite image in the suburban area, Japan. 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.

Takumi Ichimura, Shin Kamada

AffectNet contains more than 1,000,000 facial images which manually annotated for the presence of eight discrete facial expressions and the intensity of valence and arousal. Adaptive structural learning method of DBN (Adaptive DBN) is positioned as a top Deep learning model of classification capability for some large image benchmark databases. The Convolutional Neural Network and Adaptive DBN were trained for AffectNet and classification capability was compared. Adaptive DBN showed higher classification ratio. However, the model was not able to classify some test cases correctly because human emotions contain many ambiguous features or patterns leading wrong answer which includes the possibility of being a factor of adversarial examples, due to two or more annotators answer different subjective judgment for an image. 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. In addition, an appropriate child model was generated according to difference between two models by using KL divergence. The generated child models showed better performance to classify two emotion categories: `Disgust' and `Anger'.

Shin Kamada, Takumi Ichimura

Deep learning builds deep architectures such as multi-layered artificial neural networks to effectively represent multiple features of input patterns. The adaptive structural learning method of Deep Belief Network (DBN) can realize a high classification capability while searching the optimal network structure during the training. 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. Moreover, the learning algorithm of Adaptive RBM and Adaptive DBN was extended to the time-series analysis by using the idea of LSTM (Long Short Term Memory). In this paper, our proposed prediction method was applied to Moving MNIST, which is a benchmark data set for video recognition. We challenge to reveal the power of our proposed method in the video recognition research field, since video includes rich source of visual information. Compared with the LSTM model, our method showed higher prediction performance (more than 90% predication accuracy for test data).

Shin Kamada, Takumi Ichimura

Deep learning forms a hierarchical network structure for representation of multiple input features. The adaptive structural learning method of Deep Belief Network (DBN) can realize a high classification capability while searching the optimal network structure during the training. The method can find the optimal number of hidden neurons for given input data in a Restricted Boltzmann Machine (RBM) by neuron generation-annihilation algorithm. Moreover, it can generate a new hidden layer in DBN by the layer generation algorithm to actualize a deep data representation. The proposed method showed higher classification accuracy for image benchmark data sets than several deep learning methods including well-known CNN methods. In this paper, a new object detection method for the DBN architecture is proposed for localization and category of objects. The method is a task for finding semantic objects in images as Bounding Box (B-Box). To investigate the effectiveness of the proposed method, the adaptive structural learning of DBN and the object detection were evaluated on the Chest X-ray image benchmark data set (CXR8), which is one of the most commonly accessible radio-logical examination for many lung diseases. The proposed method showed higher performance for both classification (more than 94.5% classification for test data) and localization (more than 90.4% detection for test data) than the other CNN methods.

Shin Kamada, Takumi Ichimura, Toshihide Harada

Deep Learning has a hierarchical network architecture to represent the complicated feature of input patterns. The adaptive structural learning method of Deep Belief Network (DBN) has been developed. The method can discover an optimal number of hidden neurons for given input data in a Restricted Boltzmann Machine (RBM) by neuron generation-annihilation algorithm, and generate a new hidden layer in DBN by the extension of the algorithm. In this paper, the proposed adaptive structural learning of DBN was applied to the comprehensive medical examination data for the cancer prediction. The prediction system shows higher classification accuracy (99.8% for training and 95.5% for test) than the traditional DBN. Moreover, the explicit knowledge with respect to the relation between input and output patterns was extracted from the trained DBN network by C4.5. Some characteristics extracted in the form of IF-THEN rules to find an initial cancer at the early stage were reported in this paper.

Shin Kamada, Takumi Ichimura

Recently, the market on deep learning including not only software but also hardware is developing rapidly. Big data is collected through IoT devices and the industry world will analyze them to improve their manufacturing process. Deep Learning has the hierarchical network architecture to represent the complicated features of input patterns. Although deep learning can show the high capability of classification, prediction, and so on, the implementation on GPU devices are required. We may meet the trade-off between the higher precision by deep learning and the higher cost with GPU devices. We can success the knowledge extraction from the trained deep learning with high classification capability. The knowledge that can realize faster inference of pre-trained deep network is extracted as IF-THEN rules from the network signal flow given input data. Some experiment results with benchmark tests for time series data sets showed the effectiveness of our proposed method related to the computational speed.

Takumi Ichimura, Shin Kamada

Deep Learning has the hierarchical network architecture to represent the complicated features of input patterns. Such architecture is well known to represent higher learning capability compared with some conventional models if the best set of parameters in the optimal network structure is found. We have been developing the adaptive learning method that can discover the optimal network structure in Deep Belief Network (DBN). The learning method can construct the network structure with the optimal number of hidden neurons in each Restricted Boltzmann Machine and with the optimal number of layers in the DBN during learning phase. The network structure of the learning method can be self-organized according to given input patterns of big data set. In this paper, we embed the adaptive learning method into the recurrent temporal RBM and the self-generated layer into DBN. In order to verify the effectiveness of our proposed method, the experimental results are higher classification capability than the conventional methods in this paper.

Shin Kamada, Takumi Ichimura

Recently, Deep Learning has been applied in the techniques of artificial intelligence. Especially, Deep Learning performed good results in the field of image recognition. Most new Deep Learning architectures are naturally developed in image recognition. For this reason, not only the numerical data and text data but also the time-series data are transformed to the image data format. Multi-modal data consists of two or more kinds of data such as picture and text. The arrangement in a general method is formed in the squared array with no specific aim. In this paper, the data arrangement are modified according to the similarity of input-output pattern in Adaptive Structural Learning method of Deep Belief Network. The similarity of output signals of hidden neurons is made by the order rearrangement of hidden neurons. The experimental results for the data rearrangement in squared array showed the shortening time required for DBN learning.

Shin Kamada, Takumi Ichimura, Takanobu Watanabe

The recommendation system of the competitive grants to university researchers by using the Grants-in-Aid for Scientific Research (KAKEN) keywords has been developed. The system can determine the recommendation order of researchers to each grant by the using the association rules between KAKEN application and various information from the web site of the corresponding grant. However, our developed previous system has some fatal errors in the retrieval algorithm. We modify the algorithm and extend the retrieval data for web mining. If the grant information is not enough to determine the relation, the system investigates the past KAKEN records in the database for the researcher who acquired the past grant. Moreover, the system retrieves the papers of the researchers to search their interests. As a result, the agreement degree of the researcher's interest to the grant increases. This paper discusses some simulation results.

Takumi Ichimura, Takuya Uemoto, Shin Kamada

We have already developed the recommendation system of sightseeing information on SNS by using smartphone based user participatory sensing system. The system can post the attractive information for tourists to the specified Facebook page by our developed smartphone application. The users in Facebook, who are interested in sightseeing, can come flocking through information space from far and near. However, the activities in the community on SNS are only supported by the specified people called a hub. We proposed the method of vitalization of tourist behaviors to give a stimulus to the people. We developed the simulation system for multi agent system with altruistic behaviors inspired by the Army Ants. The army ant takes feeding action with altruistic behaviors to suppress selfish behavior to a common object used by a plurality of users in common. In this paper, we introduced the altruism behavior determined by some simulation to vitalize the SNS community. The efficiency of the revitalization process of the community was investigated by some experimental simulation results.

Shin Kamada, Takumi Ichimura

We developed an adaptive structure learning method of Restricted Boltzmann Machine (RBM) which can generate/annihilate neurons by self-organizing learning method according to input patterns. Moreover, the adaptive Deep Belief Network (DBN) in the assemble process of pre-trained RBM layer was developed. The proposed method presents to score a great success to the training data set for big data benchmark test such as CIFAR-10. However, the classification capability of the test data set, which are included unknown patterns, is high, but does not lead perfect correct solution. We investigated the wrong specified data and then some characteristic patterns were found. In this paper, the knowledge related to the patterns is embedded into the classification algorithm of trained DBN. As a result, the classification capability can achieve a great success (97.1\% to unknown data set).

Shin Kamada, Takumi Ichimura

Deep Belief Network (DBN) has a deep architecture that represents multiple features of input patterns hierarchically with the pre-trained Restricted Boltzmann Machines (RBM). A traditional RBM or DBN model cannot change its network structure during the learning phase. Our proposed adaptive learning method can discover the optimal number of hidden neurons and weights and/or layers according to the input space. The model is an important method to take account of the computational cost and the model stability. The regularities to hold the sparse structure of network is considerable problem, since the extraction of explicit knowledge from the trained network should be required. In our previous research, we have developed the hybrid method of adaptive structural learning method of RBM and Learning Forgetting method to the trained RBM. In this paper, we propose the adaptive learning method of DBN that can determine the optimal number of layers during the learning. We evaluated our proposed model on some benchmark data sets.

Shin Kamada, Takumi Ichimura

Restricted Boltzmann Machine (RBM) is a generative stochastic energy-based model of artificial neural network for unsupervised learning. Recently, RBM is well known to be a pre-training method of Deep Learning. In addition to visible and hidden neurons, the structure of RBM has a number of parameters such as the weights between neurons and the coefficients for them. Therefore, we may meet some difficulties to determine an optimal network structure to analyze big data. In order to evade the problem, we investigated the variance of parameters to find an optimal structure during learning. For the reason, we should check the variance of parameters to cause the fluctuation for energy function in RBM model. In this paper, we propose the adaptive learning method of RBM that can discover an optimal number of hidden neurons according to the training situation by applying the neuron generation and annihilation algorithm. 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. Moreover, the inactivated hidden neuron will be annihilated if the neuron does not affect the learning situation. The experimental results for some benchmark data sets were discussed in this paper.

Shin Kamada, Takumi Ichimura, Takanobu Watanabe

An acquisition of a research grant is important for the researchers to conduct a research. The university will build up the organization and reinforce the acquirement of external funds. The researcher becomes aware of grant information and should investigate what kinds of grant it is. Therefore, the staff at the support center for the Industry-Academia collaboration will classify the grant into some categories according to the research fields. However, the task is difficult to realize the matching of the research fields, because the expert knowledge is required to completely classify them. We have developed recommendation system of Grant-in-Aid system for researchers by using JSPS (Japan Society for the Promotion of Science) keywords. The characteristic keywords are extracted from web sites and then the association rules between researchers and grants are determined in the IF-THEN rule format. This paper discusses the experimental results by using the developed system.

Shin Kamada, Takumi Ichimura

Recently, a high technique of image processing is required to extract the image features in real time. In our research, the tourist subject data are collected from the Mobile Phone based Participatory Sensing (MPPS) system. Each record consists of image files with GPS, geographic location name, user's numerical evaluation, and comments written in natural language at sightseeing spots where a user really visits. In our previous research, the famous landmarks in sightseeing spot can be detected by Clonal Selection Algorithm with Immunological Memory Cell (CSAIM). However, some landmarks was not detected correctly by the previous method because they didn't have enough amount of information for the feature extraction. In order to improve the weakness, we propose the generation method of immunological memory by Restricted Boltzmann Machines. To verify the effectiveness of the method, some experiments for classification of the subjective data are executed by using machine learning tools for Deep Learning.

Takumi Ichimura, Shin Kamada

Near Field Communication (NFC) standards cover communications protocols and data exchange formats. They are based on existing radio-frequency identification (RFID) standards. In Japan, Felica card is a popular way to identify the unique ID. Recently, the attendance management system (AMS) with RFID technology has been developed as a part of Smart University, which is the educational infrastructure using high technologies, such as ICT. However, the reader/writer for Felica is too expensive to build the AMS. NFC technology includes not only Felica but other type of IC chips. The Android OS 2.3 and the later can provide access to NFC functionality. Therefore, we developed AMS for university with NFC on Nexus 7. Because Nexus 7 is a low cost smart tablet, a teacher can determine to use familiarly. Especially, this paper describes the method of early discovery for chronic non-attenders by using the AMS system on 2 or more Nexus 7 which is connected each other via peer-to-peer communication. The attendance situation collected from different Nexus 7 is merged into a SQLite file and then, the document is reported to operate with the trunk system in educational affairs section.

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. Sensed data can also be obtained from external sensing devices that can communicate wirelessly to the phone. Our developed tourist subjective data collection system with Android smartphone can determine the filtering rules to provide the important information of sightseeing spot. The rules are automatically generated by Interactive Growing Hierarchical SOM. However, the filtering rules related to photograph were not generated, because the extraction of the specified characteristics from images cannot be realized. We propose the effective method of the Levenshtein distance to deduce the spatial proximity of image viewpoints and thus determine the specified pattern in which images should be processed. To verify the proposed method, some experiments to classify the subjective data with images are executed by Interactive GHSOM and Clonal Selection Algorithm with Immunological Memory Cells in this paper.

Takumi Ichimura, Shin Kamada

The clonal selection principle explains the basic features of an adaptive immune response to a antigenic stimulus. It established the idea that only those cells that recognize the antigens are selected to proliferate and differentiate. This paper explains a computational implementation of the clonal selection principle that explicitly takes into account the affinity maturation of the immune response. Antibodies generated by the clonal selection algorithm are clustered in some categories according to the affinity maturation, so that immunological memory cells which respond to the specified pathogen are created. Experimental results to classify the medical database of Coronary Heart Disease databases are reported. For the dataset, our proposed method shows the 99.6\% classification capability of training data.