This topic created in 2689 days ago, the information mentioned may be changed or developed.
52 replies • 2019-02-21 09:40:54 +08:00
 |
1
wanmafeiteng OP BGM 带感
|
 |
2
takato Jan 25, 2019
刚搬好了小板凳。。
吐槽下 tag：deepmind StarCraft demonstration “吗” |
 |
3
tyrealgray Jan 25, 2019 via Android
已经吃着晚饭端着笔记本在边上等着了，😄
|
 |
4
Ediacaran Jan 25, 2019
差点忘了，感谢提醒
|
|
5
wanmafeiteng OP @tyrealgray 现在才吃-_-||
|
|
6
wanmafeiteng OP @Ediacaran | ू•ૅω•́)ᵎᵎᵎ
|
|
7
tyrealgray Jan 25, 2019 via Android
@wanmafeiteng 欧洲，我这还没到六点
|
|
8
wanmafeiteng OP @takato →_→
|
|
9
wanmafeiteng OP 开始喽～(￣▽￣～)~
|
 |
10
shadiao Jan 25, 2019 via iPhone
@wanmafeiteng 建群了吗？
|
|
11
tyrealgray Jan 25, 2019 via Android
为啥种族是定了的，不应该随机吗？
|
|
12
takato Jan 25, 2019
建筑居然造基地旁边，好奇怪的思路啊= =
|
 |
13
t6attack Jan 25, 2019
墙内中文转播 https://www.douyu.com/534103
|
|
14
wanmafeiteng OP 人类第一局输了
|
|
15
tyrealgray Jan 25, 2019 via Android
等等，这是录播啊
|
|
16
wanmafeiteng OP @tyrealgray T_T
|
|
17
wanmafeiteng OP @shadiao 没有
|
|
18
wanmafeiteng OP 人类真惨-_-||
|
 |
19
yaoppp Jan 25, 2019 via iPhone
AI 上来狂造农民
AI 的 apm 平均 277，人类选手是 559 |
|
20
takato Jan 25, 2019
自爆球无敌。。。- -
|
|
21
wanmafeiteng OP 人类输三局-_-||
|
|
22
wanmafeiteng OP @takato 的确猛(ಡωಡ)
|
 |
23
fe619742721 Jan 25, 2019 via iPhone
TLO 被老仙喷惨了，怎么选了个这么菜的人类选手上去，打起来还像是收了钱的。。。
老仙预计接班人都能打过 ai 哈哈 |
|
24
wanmafeiteng OP @fe619742721 在哪？
|
 |
25
scnace Jan 25, 2019 via Android
TLO 真的给人类丢脸
|
 |
26
Xs0ul Jan 25, 2019
一切战术转换家
|
|
27
tyrealgray Jan 25, 2019
输的太惨了
|
|
28
wanmafeiteng OP @tyrealgray 崩
|
|
29
wanmafeiteng OP 人类太烂了
|
|
30
wanmafeiteng OP 追猎流
|
 |
31
goldkeyber Jan 25, 2019 via Android
这么看的话，the little one 不是演就是真的菜，两周比一周的进步太大了，mana 打不过是真的打不过。。
|
|
32
wanmafeiteng OP wc，能赢
|
|
33
wanmafeiteng OP 人类赢了
|
|
34
goldkeyber Jan 25, 2019 via Android
阴险的人类。。
|
|
35
tyrealgray Jan 25, 2019
我去，为啥赢的这一局 AI 根本没有反抗
|
|
36
wanmafeiteng OP @tyrealgray 人工智能太傻了
|
|
37
wanmafeiteng OP 睡觉(＠￣ー￣＠)
|
 |
38
alphanow Jan 25, 2019
Live 的 AI 和之前的录比赛的不一样，录比赛时候的 AI 是用的全图模式（还是有战争迷雾的），Live 的是正式游戏的那种只能看到一屏视野的
|
|
39
wanmafeiteng OP AI 好像不会对战中同时学习
|
|
40
alphanow Jan 25, 2019
@wanmafeiteng 对战是验证过程不是学习过程
|
|
41
tyrealgray Jan 25, 2019
@goldkeyber TLO 主打的虫族，采访有说他如果是用虫族有很大把握赢
|
 |
42
jasonyang9 Jan 25, 2019
请教具体在哪里看的？斗鱼上没找到录播
|
|
44
alphanow Jan 25, 2019 via Android
@jasonyang9 deepmind 官网有全部 11 场比赛的 replay
|
|
45
jasonyang9 Jan 25, 2019
@alphanow #44 找到了。问题是我没 SC2，囧
|
 |
46
Tyanboot PRO @alphanow 之前的也不是全图视野吧。TLO 第一个 replay 放完之后给了个 AI 视角演示，可以看到有限制视野和切屏操作。
|
|
47
wanmafeiteng OP AI 能不能做到自己假设情况去进行思考🤔
|
 |
48
ballshapesdsd Jan 25, 2019
谁能告诉我，为啥楼主的回帖都是高亮的
|
|
49
alphanow Jan 25, 2019
@Tyanboot 那个是操作视野，我复制了 Deepmind 的一段 blog 过来
>During the matches against TLO and MaNa, AlphaStar interacted with the StarCraft game engine directly via its raw interface, meaning that it could observe the attributes of its own and its opponent ’ s visible units on the map directly, without having to move the camera - effectively playing with a zoomed out view of the game. In contrast, human players must explicitly manage an "economy of attention" to decide where to focus the camera. However, analysis of AlphaStar ’ s games suggests that it manages an implicit focus of attention. On average, agents “ switched context ” about 30 times per minute, similar to MaNa or TLO. > >Additionally, and subsequent to the matches, we developed a second version of AlphaStar. Like human players, this version of AlphaStar chooses when and where to move the camera, its perception is restricted to on-screen information, and action locations are restricted to its viewable region. |
 |
50
nicevar Jan 25, 2019
AI 的操作和运营能做到无敌，人类要靠偷鸡才行，过段时间就完全干不过了
|
|
51
wanmafeiteng OP @jasonyang9 deempind 官网有录像
|
 |
52
AsunaQAQ Feb 21, 2019
当时解说 我注意到的一点就是 打架可以让兵营无歇息亮灯 我记得
|
Select Language