A long-standing goal of artificial intelligence is an algorithm that learns, tabula rasa, superhuman proficiency in challenging domains. Recently, AlphaGo became the first program to defeat a world ...champion in the game of Go. The tree search in AlphaGo evaluated positions and selected moves using deep neural networks. These neural networks were trained by supervised learning from human expert moves, and by reinforcement learning from self-play. Here we introduce an algorithm based solely on reinforcement learning, without human data, guidance or domain knowledge beyond game rules. AlphaGo becomes its own teacher: a neural network is trained to predict AlphaGo's own move selections and also the winner of AlphaGo's games. This neural network improves the strength of the tree search, resulting in higher quality move selection and stronger self-play in the next iteration. Starting tabula rasa, our new program AlphaGo Zero achieved superhuman performance, winning 100-0 against the previously published, champion-defeating AlphaGo.
The game of Go has long been viewed as the most challenging of classic games for artificial intelligence owing to its enormous search space and the difficulty of evaluating board positions and moves. ...Here we introduce a new approach to computer Go that uses 'value networks' to evaluate board positions and 'policy networks' to select moves. These deep neural networks are trained by a novel combination of supervised learning from human expert games, and reinforcement learning from games of self-play. Without any lookahead search, the neural networks play Go at the level of state-of-the-art Monte Carlo tree search programs that simulate thousands of random games of self-play. We also introduce a new search algorithm that combines Monte Carlo simulation with value and policy networks. Using this search algorithm, our program AlphaGo achieved a 99.8% winning rate against other Go programs, and defeated the human European Go champion by 5 games to 0. This is the first time that a computer program has defeated a human professional player in the full-sized game of Go, a feat previously thought to be at least a decade away.
Go nation Moskowitz, Marc L
2013., 20130831, 2013, 2013-08-31, Letnik:
28
eBook
Go (Weiqi in Chinese) is one of the most popular games in East Asia, with a steadily increasing fan base around the world. Like chess, Go is a logic game but it is much older, with written records ...mentioning the game that date back to the 4th century BC. As Chinese politics have changed over the last two millennia, so too has the imagery of the game. In Imperial times it was seen as a tool to seek religious enlightenment and was one of the four noble arts that were a requisite to becoming a cultured gentleman. During the Cultural Revolution it was a stigmatized emblem of the lasting effects of feudalism. Today, it marks the reemergence of cultured gentlemen as an idealized model of manhood. Marc L. Moskowitz explores the fascinating history of the game, as well as providing a vivid snapshot of Chinese Go players today. Go Nation uses this game to come to a better understanding of Chinese masculinity, nationalism, and class, as the PRC reconfigures its history and traditions to meet the future.
Objectives
The purpose of this study is to clarify the influence of social interaction on the effect of a cognitive intervention program using Go.
Methods
A single‐blind, randomized controlled trial ...using a classical board game “Go” was conducted. A total of 72 community‐dwelling older adults, without previous experience playing Go, were randomly assigned to three groups: (1) a face‐to‐face group (FG) in which members attended 12 Go group lessons held once a week; (2) a non‐face‐to‐face group (NFG) in which members individually underwent the same Go lessons as the FG using a tablet computer; or (3) a health education control group (CG). The main outcome variable, working memory, was assessed before and after the interventions using the Visual Memory Span Test (VMST) and the Visual Memory Span Backward (VMSB) task. Go performance and additional cognitive domains were also examined.
Results
Analysis of covariance revealed that VMST scores significantly improved after the intervention in both the FG and NFG (both P < .05). Compared with the CG, the effect size of the FG (Cohen's d = 0.89) was greater than that of the NFG (Cohen's d = 0.67). Although VMSB scores significantly improved after the intervention in the FG (P < .05), no significant changes were observed in other groups.
Conclusions
This study showed that Go game could improve visual working memory regardless of social interaction. Furthermore, findings suggested that playing board games face‐to‐face with others is more effective for cognitive function than playing alone.
Background
Influence of cognitive intervention programs on brain activity has not been enough explored.
Aims
The aims of the present study were to clarify changes in brain activity from a cognitive ...intervention program utilizing the board game “Go” and to examine the relationship between brain activity and the acquisition of Go skills.
Methods
Eighteen community-dwelling older adults were randomly assigned either to an intervention group (IG), in which members attended 12 Go lessons either in groups or individually using tablet computers, or a control group (CG), in which members attended health education lectures unrelated to Go. Fluorine-18 fluorodeoxyglucose (
18
F-FDG) positron emission tomography (PET), cognitive assessments, and Go tests were performed before and after the intervention.
Results
The results showed different patterns of regional FDG uptake in both groups: regional cerebral glucose metabolism was significantly increased in the left middle temporal gyrus (MTG) and bilateral putamen (
p
< 0.01; cluster level) in the IG, and in the left superior frontal gyrus in the CG, (
p
< 0.01; cluster level). Furthermore, Go test scores were significantly improved in the IG (
p
< 0.05), and a significant association was observed between changes in Go test scores and glucose metabolism in the left MTG (
p
< 0.05).
Conclusions and discussion
This study indicates that a cognitive intervention program using Go may enhance brain activity. Further studies with larger populations and longer observation periods are needed to clarify the neural mechanisms underlying our Go intervention program.
Go is a game that can be won or lost based on the number of intersections surrounded by black or white pieces. The traditional method is a manual counting method, which is time-consuming and ...error-prone. In addition, the generalization of the current Go-image-recognition methods is poor, and accuracy needs to be further improved. To solve these problems, a Go-game image recognition based on an improved pix2pix was proposed. Firstly, a channel-coordinate mixed-attention (CCMA) mechanism was designed by combining channel attention and coordinate attention effectively; therefore, the model could learn the target feature information. Secondly, in order to obtain the long-distance contextual information, a deep dilated-convolution (DDC) module was proposed, which densely linked the dilated convolution with different dilated rates. The experimental results showed that compared with other existing Go-image-recognition methods, such as DenseNet, VGG-16, and Yolo v5, the proposed method could effectively improve the generalization ability and accuracy of a Go-image-recognition model, and the average accuracy rate was over 99.99%.
PDF
