Christopher D. Rosin

From Chessprogramming wiki
Jump to: navigation, search

Home * People * Christopher D. Rosin

Christopher D. Rosin [1]

Christopher Darrell (Chris) Rosin,
an American computer scientist working in the fields of machine learning and medical monitoring, as well as computer Go programmer. He defended his Ph.D. in 1997 at University of California, San Diego on the topic of genetic algorithms [2].

PUCT

As author of the Go playing program Greenpeep [3], Chris Rosin devised an improvement of UCT dubbed PUCT, as implemented in his program, the idea further published and elaborated in Multi-armed bandits with episode context [4]. PUCT modifies the original UCB1 multi-armed bandit policy by approximately predicting good arms at the start of a sequence of multi-armed bandit trials ('Predictor' + UCB = PUCB). A variation of PUCT was used in the AlphaGo and AlphaZero projects [5] , and subsequently also in Leela Zero and Leela Chess Zero [6].

Selected Publications

[7] [8]

1995 ...

2000 ...

2010 ...

External Links

References

Up one level