FML-based Dynamic Assessment Agent for Human-Machine Cooperative System on Game of Go

Abstract

In this paper, we demonstrate the application of Fuzzy Markup Language (FML) to construct an FML-based Dynamic Assessment Agent (FDAA), and we present an FML-based Human-Machine Cooperative System (FHMCS) for the game of Go. The proposed FDAA comprises an intelligent decision-making and learning mechanism, an intelligent game bot, a proximal development agent, and an intelligent agent. The intelligent game bot is based on the open-source code of Facebook's Darkforest, and it features a representational state transfer application programming interface mechanism. The proximal development agent contains a dynamic assessment mechanism, a GoSocket mechanism, and an FML engine with a fuzzy knowledge base and rule base. The intelligent agent contains a GoSocket engine and a summarization agent that is based on the estimated win rate, real-time simulation number, and matching degree of predicted moves.

FML-based prediction agent and its application to game of Go

“…3 (a)-3 (b) show the predicted next moves as well as history information like winning rate, the number of simulations, and each-move inferred current game situation provided by the machine recommendation platform. FML assessment engine consists of four different methods, including winning rate difference, simulation difference, FML Assessment-1, and FML Assessment-2 [18]. In Fig.…”

“… Stage II: FML assessment engine adopts each-moveposition, darkforest-predicted top-5-move number of simulations and winning rate to decide each-move number of simulations, winning rate, and top-move rate (TMR) [18]. …”

“…It also reports the real-time predicted top-3-move positions to the human Go player to think carefully before playing his /his next move [18].…”

“…In this paper, in addition to information provided by darkforest program, we have an additional system to show the information of the game such as current game situation. According to the first-stage prediction results of darkforest Go engine [8,9] and the second-stage inferred results of the FML assessment engine [18], we further introduce the third-stage FML-based decision support engine to predict the winner of the game. Next, we choose seven games from 60 games (Master vs. top professional Go players in Dec. 2016 and in Jan. 2017) and three games of amateur Go players to evaluate the performance.…”

Ontology-based Fuzzy Markup Language Agent for Student and Robot Co-Learning

“…Fuzzy Markup Language (FML) is a specific purpose markup language based on XML to describe the structure and behavior of a fuzzy system independently of the hardware architecture [13][14]. Since May 2016, FML has become one of the IEEE standards [15][16] and has been applied to a lot of researches like game of Go [11,17], diet assessment [14,18,22], and so on. Nowadays, human and machine co-learning is an important topic for current societies.…”