2016年3月14日 星期一

Deep Blue developer speaks on how to beat Go and... 「蒙地卡羅樹搜尋」(Monte Carlo Tree Search)與「樹狀結構信賴上界法」(Upper Confidence bounds applied to Trees)


Deep Blue developer speaks on how to beat Go and crack ...

www.theverge.com/.../ibm-deep-blue-murray-campbell-alphago-deepmi...
2 days ago - IBM's supercomputer beat Garry Kasparov nearly 20 years ago .... a general mechanism, had it been created in AlphaGo and applied to chess, ...



黃士傑 Alphago

電腦圍棋的研究開始於1970年,但圍棋程式卻從未曾被人們認為是強大的,直到2006年,當「蒙地卡羅樹搜尋」(Monte Carlo Tree Search)與「樹狀結構信賴上界法」(Upper Confidence bounds applied to Trees)出現之後,情況才開始完全不同。「蒙地卡羅樹搜尋」與「樹狀結構信賴上界法」所帶進的革命強而有力到一個地步,人們甚至開始相信,圍棋程式在10年或者20年之後,將能夠擊敗頂尖的人類棋手。
ETDS.LIB.NTNU.EDU.TW

2016年3月1日 星期二

Ability to ignore is a key part of paying attention,

Ability to ignore is a key part of paying attention, Johns Hopkins study finds

Jill Rosen /  February 27
Johns Hopkins University
Latest research: People searching for something can find it faster if they know what to look for. But new research suggests knowing what not to look for can be just as helpful.