Artificial Intelligence for Communications and Networks. 4th EAI International Conference, AICON 2022 Hiroshima, Japan, November 30 – December 1, 2022 Proceedings

Preface
Organization
Contents
AI and Networks
Cost-Aware Node Ranking Algorithm for Embedding Virtual Networks in Internet of Vehicles
	1 Introduction
	2 Related Work
	3 Network Model and Problem Description
		3.1 Virtual Network Assignment
		3.2 Performance Metrics
	4 Proposed Solution
		4.1 Pre-processing Stage
		4.2 Vehicle Ranking
		4.3 Intermediate Node Cost
		4.4 VNE-IoV Algorithm
	5 Performance Evaluation
		5.1 Simulation Setup
		5.2 Vehicle Mobility Model
		5.3 Evaluation Results
	6 Conclusion and Future Work
	References
Fault Diameter of Strong Product Graph of Two Paths
	1 Introduction
	2 Main Results
	3 Model Comparison
	4 Conclusions
	References
Design and Implementation of SF Selection Based on Distance and SNR Using Autonomous Distributed Reinforcement Learning in LoRa Networks
	1 Introduction
	2 System Model and Problem Formulation
	3 SF Parameter Selection for ToW-Dynamics
		3.1 Multi-armed Bandit Problem
		3.2 ToW-Dynamics Based SF Selection
	4 Implementation and Performance Evaluation
	5 Conclusion
	References
Machine Learning
QBRT: Bias and Rising Threshold Algorithm with Q-Learning
	1 Introduction
	2 Related Studies
		2.1 Prior Work on Environment Non-stationarity and Scalability Issues
		2.2 BRT Algorithm
		2.3 Best-of-n Problem
		2.4 Q Learning
	3 Proposed Method
		3.1 QBRT
		3.2 Behavior of Swarms Implementing QBRT
	4 Experiments
		4.1 Experimental Environment
		4.2 Experiment at THM with p=3,d=3
		4.3 Experiment at THM with p=3,d=5
	5 Conclusion
	References
A Study on Effectiveness of BERT Models and Task-Conditioned Reasoning Strategy for Medical Visual Question Answering
	1 Introduction
	2 Related Work
		2.1 VQA-RAD Dataset
		2.2 CMSA-MTPT Framework
		2.3 Task-Condition Reasoning Strategy
	3 Our Framework
		3.1 Overview
		3.2 BERT Models as the Language Understanding Module
		3.3 Setting of the Task-Conditioned Strategy
	4 Experiments
		4.1 Training ResNet Models
		4.2 CMSA-MTPT with BERT Models
		4.3 Task-Conditioned Reasoning CMSA-MTPT with BERT Models
		4.4 A Clue for Interpretability in BERT Models
	5 Conclusion
	References
Deep Robust Neural Networks Inspired by Human Cognitive Bias Against Transfer-based Attacks
	1 Introduction
	2 Related Work
		2.1 Adversarial Examples
		2.2 Transfer-Based Attacks
		2.3 Defense Against Transfer-Based Attacks
	3 Proposed Method
		3.1 Loosely Symmetric Neural Networks
		3.2 LS-DNN Development
	4 Experiments and Discussion
		4.1 Methods
		4.2 Conditions
		4.3 Results
	5 Conclusion and Future Work
	References
Efficient Estimation of Cow\'s Location Using Machine Learning Based on Sensor Data
	1 Introduction
	2 Related Works
	3 Methods
		3.1 Data Acquisition
		3.2 Data Preprocessing
		3.3 Machine Learning Model
	4 Results and Discussion
	5 Conclusions
	References
A Time Series Forecasting Method Using DBN and Adam Optimization
	1 Introduction
	2 DBN for Time Series Forecasting
		2.1 RBM and Its Learning Rule
		2.2 MLP and Its Learning Rule
		2.3 Meta Parameter Optimization
	3 Experiments and Analysis
		3.1 Benchmark CATS
		3.2 Results and Analysis of CATS Forecasting
		3.3 Chaotic Time Series Data
		3.4 Results and Analysis of Chaotic Time Series Forecasting
	4 Conclusions
	References
Unity-Bounded Function and Benchmark Design Specifications Targeted for Designing Typical Variable Digital Filters
	1 Introduction
	2 Typical Benchmark Specifications
		2.1 Variable Lowpass Design Specification
		2.2 Variable Highpass Design Specification
		2.3 Variable Bandpass Design Specification
		2.4 Variable Bandstop Design Specification
		2.5 Variable Notch-Frequency Design Specification
	3 Unity-Bounded Function
	4 An Illustrative Example
	5 Conclusion
	References
Evolutionary Computation
Proposal and Evaluation of a Course-Classification-Support System Emphasizing Communication with the Sub-committees Within the Committee of Validation and Examination for Degrees
	1 Introduction
	2 Degree-Awarding of NIAD-QE and the Course-Classification-Support System
	3 The Course-Classification-Support System Using the Deep Learning
		3.1 Preparation of Training Data and Previous Methods
		3.2 Relationship with Related Works and Our Approach
		3.3 Proposed Method
	4 Results and Discussion
	5 Conclusions
	References
A Research of Infectivity Rate of Seasonal Influenza from Pre-infectious Person for Data Driven Simulation
	1 Introduction
	2 Proposed SEPIR Model
	3 Epidemic Situations
		3.1 Situation A
		3.2 Situation B
		3.3 Situation C
	4 Parameters
		4.1 Infectivity Rate
		4.2 The Number of Susceptible Students
	5 Near Decomposability
		5.1 Definition
		5.2 Rough Estimation
	6 Discussion
	7 Conclusion
	References
Creating Trust Within Population of Evolutionary Computation in an Uncertain Environment Using Blockchain
	1 Introduction
	2 Usecases of Blockchain in Evolutionary Computation
	3 Trust in Evolutionary Computation with Blockchain
	4 System Architecture
	5 Conclusion
	References
Efficient Inductive Logic Programming Based on Particle Swarm Optimization
	1 Introduction
	2 Progol
	3 Particle Swarm Optimization
	4 PSO Based Progol
	5 Experiments
	6 Conclusions and Future Work
	References
Author Index