Abstract
In this paper, we propose an AI-FML agent for robotic game of Go and AIoT real-world co-learning applications. The fuzzy machine learning mechanisms are adopted in the proposed model, including fuzzy markup language (FML)-based genetic learning (GFML), eXtreme Gradient Boost (XGBoost), and a seven-layered deep fuzzy neural network (DFNN) with backpropagation learning, to predict the win rate of the game of Go as Black or White. This paper uses Google AlphaGo Master sixty games as the dataset to evaluate the performance of the fuzzy machine learning, and the desired output dataset were predicted by Facebook AI Research (FAIR) ELF Open Go AI bot. In addition, we use IEEE 1855 standard for FML to describe the knowledge base and rule base of the Open Go Darkforest (OGD) prediction platform in order to infer the win rate of the game. Next, the proposed AI-FML agent publishes the inferred result to communicate with the robot Kebbi Air based on MQTT protocol to achieve the goal of human and smart machine co-learning. From Sept. 2019 to Jan. 2020, we introduced the AI-FML agent into the teaching and learning fields in Taiwan. The experimental results show the robots and students can co-learn AI tools and FML applications effectively. In addition, XGBoost outperforms the other machine learning methods but DFNN has the most obvious progress after learning. In the future, we hope to deploy the AI-FML agent to more available robot and human co-learning platforms through the established AI-FML International Academy in the world.
-
IEEE Keywords
- Games,
- Robots,
- Machine learning,
- Protocols,
- Fuzzy logic,
- Knowledge-based systems
-
Author Keywords
- Intelligent Agent,
- Robot,
- Fuzzy Machine Learning,
- Game of Go ,
- AIoT
Introduction
With the success of AlphaGo [1], there has been a lot of interest among students and professionals to apply machine learning to gaming and in particular to the game of Go. Go is a highly competitive and time-consuming activity and each move played on the Go board can strongly make an influence on the probability of winning or losing the game [4]. The Open Go Darkforest (OGD) prediction platform has been constructed since 2016 [3, 4, 5, 6]. The OGD cloud platform, including a Facebook AI Research (FAIR) dynamic Darkforest (DDF) AI bot and a FAIR ELF Open Go AI bot prediction mechanisms, has the ability to predict the top five selections for the next move. The AI bot provides each selection with real-time win rate, simulation numbers, and the top-move matching rate [6]. The IEEE Computational Intelligence Society (CIS) funded Fuzzy Markup Language (FML)-based machine learning competition for human and smart machine co-learning on game of Go in IEEE CEC 2019 and FUZZ-IEEE 2019. The goal of the competition is to understand the basic concepts of an FML-based fuzzy inference system, to use the FML intelligent decision tool to establish the knowledge base and rule base of the fuzzy inference system, and to optimize the FML knowledge base and rule base through the methodologies of evolutionary computation and machine learning [2]. Machine learning has been a hot topic in research and industry and new methodologies keep developing all the time. Regression, ensemble methods, and deep learning are important machine learning methods for data scientists [10]. Extreme gradient boost (XGBoost) is a scalable tree boosting system that provides state-of-the-art results on many problems [11]. For example, Song et al. [12] proposed an optimization model combining XGBoost algorithm with improved particle swarm optimization (PSO) to address the continuous multivariable optimization problem. Jiang et al. [13] proposed a genetic algorithm (GA)-XGBoost classifier for pedestrian detection to improve the classification accuracy. Zhang et al. [14] proposed XGBoost-based algorithm to recognize five indoor activities and its performance is better than the other ensemble learning classifier and single classifiers. Deep learning architectures have been applied to many research fields and in some cases surpass human expert performance [15]. Based on deep reinforcement learning, Ding et al. [16] proposed a novel intelligent diagnosis method to overcome the shortcomings of the diagnosis methods. Huang et. al. [17] developed a neural network scheme to extract information from emails to enable its transformation into a multidimensional vector. It is estimated that there will be 4.5 billion Internet of Things (IoTs) joining the Internet by 2020 [18]. Alaiz-Moreton et al. [19] created classification models based on ensemble methods and deep learning models to classify the attacks of an IoT system that uses the MQTT protocol. Al-Ali et al. [19] presented an energy management system for smart homes to better manage energy consumption by utilizing the technologies of IoT and big data. In 2020, in addition to Go, the organizers of the FML-based machine learning competition for human and smart machine co-learning on game of Go/AIoT at IEEE WCCI 2020 further propose the AI-FML agent to integrate AI tools, i.e., using AI- FML tools to construct the knowledge base and rule base of the fuzzy inference system, with Internet of Things (IoT). The AI-FML agent can communicate with IoT devices or robots, and it has been introduced to the learning course of 5-grade computer studies at Rende elementary school in Taiwan from Sept. 2019 o Jan. 2020. This paper uses AlphaGo Master sixty games [7] as the experimental dataset downloaded from the website of the competition @ IEEE WCCI 2020. We use the data predicted by DDF AI bot [8] as the training data, and the data predicted by ELF Open Go AI bot [9] are as the desired output of the training data. The goal of the fuzzy machine learning mechanism in this paper is to make the win rates predicted by the DDF AI bot closer to those predicted by the ELF Open Go AI bot. In addition, the proposed AI-FML agent communicates with the robot Kebbi Air through MQTT protocol to allow the students who joined the human and smart machine co-learning course to directly interact with the robot to simulate their learning motivation. The remainder of the paper is organized as follows. We first introduce the two-stage system structure for game of Go and AIoT applications in Section II. We then describe the dataset used in the FML-based machine learning competition at IEEE WCCI 2020 in Section III as well as the adopted fuzzy machine learning mechanisms in Section IV. Experimental results will be given in Section V and finally we conclude the paper in Section VI.
Conclusion
This paper proposes an AI-FML agent to integrate AI tools with IoT for real-world applications. The AI-FML agent can communicate with IoT devices or robots, and it has been introduced to the learning course of 5-grade computer studies at Rende elementary school in Taiwan from Sept. 2019 to Jan. 2020. Additionally, this paper uses AlphaGo Master sixty games as the experimental dataset to make the win rates predicted by the DDF AI bot closer to those predicted by the ELF Open Go AI bot based on FML-based genetic learning (GFML), XGBoost learning, and DFNN learning. The experimental results show that XGBoost learning and DFNN learning have the good performance as well as AI-FML agent integrating with the robot Kebbi Air is popular with the involved students. In the future, we will discuss the difference in the performance among the used learning mechanisms, expand the discussion to help provide additional insight to the readers, and deploy the AI-FML agent to more available robot and human co-learning platforms through the established AI-FML International Academy in the world.
About KSRA
The Kavian Scientific Research Association (KSRA) is a non-profit research organization to provide research / educational services in December 2013. The members of the community had formed a virtual group on the Viber social network. The core of the Kavian Scientific Association was formed with these members as founders. These individuals, led by Professor Siavosh Kaviani, decided to launch a scientific / research association with an emphasis on education.
KSRA research association, as a non-profit research firm, is committed to providing research services in the field of knowledge. The main beneficiaries of this association are public or private knowledge-based companies, students, researchers, researchers, professors, universities, and industrial and semi-industrial centers around the world.
Our main services Based on Education for all Spectrum people in the world. We want to make an integration between researches and educations. We believe education is the main right of Human beings. So our services should be concentrated on inclusive education.
The KSRA team partners with local under-served communities around the world to improve the access to and quality of knowledge based on education, amplify and augment learning programs where they exist, and create new opportunities for e-learning where traditional education systems are lacking or non-existent.
FULL Paper PDF file:
AI-FML Agent for Robotic Game of Go and AIoT Real-World Co-Learning Applications AI-FML Agent for Robotic Game of Go and AIoT Real-World Co-Learning ApplicationsBibliography
author
Year
2020
Title
AI-FML Agent for Robotic Game of Go and AIoT Real-World Co-Learning Applications
Publish in
2020 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), Glasgow, United Kingdom, 2020, pp. 1-8,
Doi
10.1109/FUZZ48607.2020.9177654.
PDF reference and original file: Click here
Somayeh Nosrati was born in 1982 in Tehran. She holds a Master's degree in artificial intelligence from Khatam University of Tehran.
Professor Siavosh Kaviani was born in 1961 in Tehran. He had a professorship. He holds a Ph.D. in Software Engineering from the QL University of Software Development Methodology and an honorary Ph.D. from the University of Chelsea.
Nasim Gazerani was born in 1983 in Arak. She holds a Master's degree in Software Engineering from UM University of Malaysia.
