قالب وردپرس درنا توس
Home https://server7.kproxy.com/servlet/redirect.srv/sruj/smyrwpoii/p2/ Technology https://server7.kproxy.com/servlet/redirect.srv/sruj/smyrwpoii/p2/ AlphaStar hungry for world domination in StarCraft II games

AlphaStar hungry for world domination in StarCraft II games



  AlphaStar hungry for world domination in StarCraft II fights

AlphaStar stones. If its very name suggests pride, it is well deserved.

AlphaStar is the AI ​​celebrity bot du jour, which tech sites testify to a battle that got an internet broadcast just recently.

"AlphaStar" was unveiled at a live stream in a show that focused on fighting StarCraft benefits. At the end of the day, AlphaStar won most games against the world's pro players. Old observers said, let's see, AlphaStar even learned from the same fight party and it may be invincible next time.

"DeepMind has an AI that beats the world's best StarCraft II players, and it's not even close," says Ryan Whitwam in ExtremeTech . And yes, he was talking about AlphaStar.

David Silber, co-founder, AlphaStar Project, DeepMind, said the team had wanted to move to upscale, accelerating the StarCraft project. What is StarCraft? As Blizzard said : "The StarCraft games have emerged as a" big challenge "for the AI ​​community, as they are the perfect environment for benchmarking progress towards issues such as planning, uncertainty management and spatial justification."

DeepMind, part of the alphabet group, takes games seriously and StarCraft, an esports challenger, is very useful for AI research. As DeepMind bloggers have said, society has been seeking "increasing complexity games that capture different intelligence elements needed to solve scientific and real problems."

"StarCraft games have emerged as a" big challenge "for the AI ​​community, as they are the perfect environment for benchmarking progress toward issues such as planning, uncertainty management, and spatial justification."

In StarCraft II, the multitasking player is forced to think both micro and macro. Players juggle big-picture and micro-level controls, said the DeepMind team, "pushing them to the limits of their strategic and tactical capabilities."

New Scientist offered specifications. "In StarCraft II, players must control armies across a terrain, they must build infrastructure, juggle short-term gains with long-term benefits. They also cannot always see the full map, so it must work on hunches and intuition rather than a fixed card. action. "

So who were the human professional players who saw the bottom beat them in their best efforts at not then it should happen?

Player names are MaNa and TLO. They are Grzegorz "MaNa" Komincz from Poland and Dario "TLO" Wunsch from Germany. MaNa and TLO are of the competitive Team Liquid, said Devin Coldewey in TechCrunch .

Many places reported the results of the fighting and StarCraft II was a winner. Chris Stokel-Walker in New Scientist reported that "In two of the best five series, DeepMinds hit AI two top-notch professionals 5-0."

DeepMind created five versions of their AI, AlphaStar, and trained them. Their training included human game footage. A deep neural network trained directly from raw game data using monitored learning and enhancement learning, DeepMind bloggers said. AlphaStar, added Coldewey, learned to watch people play first, but soon exploited his skills by playing against facets of his own.

Victor Tangermann for Futurism pronounced "The Lot's Over" and that "The machines have the upper hand" in an AlphaStar shutout. His story opener was that "The Revolutionary DeepMind of the Alphabet Artificial Intelligence, called AlphaStar, just dominated by a StarCraft II popular sci-fi strategy game by beating two of the world's best human players five games to zero." [19659004] Wired joined the clock call. Another triumph for bots. Tom Simonite found no reason to dilute the drama of "a three-hour YouTube stream where foreigners and robots fought against death."

How can the level be due to the impressive victory on the side of the AI?

  AlphaStar hungry for world domination in StarCraft II games

An observer raised actions per minute and how it speaks the weights. Although it may not always, it can perform much more "actions per Minute" than a human can, "because it is not limited by fleshy hands and buttons with buttons." Coldewey in TechCrunch : "APM is just a measurement among many who determine the outcome of a match, but it cannot hurt to be able to control a guy 20 times a second rather than two or three. "

Futurism also called AlphaStar benefits: it looked at an expanded bird's-eye view of the entire map, while players usually focus on a single area on the map.

"The version that survived the progams relied on an expanded display of the entire map, but human players tend to use this less as it is very difficult to perform many of the game's features in this view," New Scientist .


AI researchers to see if they can push some boundaries with StarCraft II


More Information :
deepmind.com/blog/alphastar-ma … y-game-starcraft-ii /

news.blizzard.com/en-gb/starcr … ii demonstration vod

© 2019 Science X Network

Citation :
AlphaStar hungry for world domination in StarCraft II matches (2019, January 26)
retrieved January 27, 2019
from https://techxplore.com/news/2019-01-alphastar-hungry-world-domination-starcraft.html

This document is subject to copyright. Besides any fair trade for private study or research, no
part can be reproduced without written permission. The content is for informational purposes only.


Source link