TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Google Research Football: A Novel Reinforcement Learning Environment

143 点作者 haditab将近 6 年前

9 条评论

lostdog将近 6 年前
If I read correctly, the agent only controls one player at a time. On offense it controls the player with the ball, and on defense it controls probably the player closest to the ball. The other players are controlled by the built-in AI. Controlling a single agent kind of takes away from the appeal of deep-RL: that entire teams can learn to coordinate in novel and optimal ways.
评论 #20137221 未加载
alexandercrohde将近 6 年前
I guess I don&#x27;t get it... What does this game have that SC2&#x2F;Dota doesn&#x27;t?<p>As far as I can tell, the main goal for reinforcement learning is to make it so that it doesn&#x27;t take 10k learning sessions to learn what a human can learn in a single session, and to make self-training without guiding scenarios feasible.
评论 #20136337 未加载
评论 #20136957 未加载
评论 #20138005 未加载
评论 #20136130 未加载
评论 #20137703 未加载
评论 #20138221 未加载
评论 #20138167 未加载
ur-whale将近 6 年前
&quot;real bayesians&quot; vs &quot;frequentists united&quot; at 0:33 in the video :D
empath75将近 6 年前
I bet that ai’s will find a lot of physics bugs to exploit early on.
milleramp将近 6 年前
Perhaps this will be used in live sports in the future. Giving real time feedback to players for optimum positioning. Would be a cool test but I still prefer to watch sports played the ‘traditional’ way.
评论 #20180933 未加载
7ewis将近 6 年前
Wonder if they use the same tech to predict the outcome of football matches. I&#x27;ve seen them show it on the Premier League games.
pitt1980将近 6 年前
Do all the players have the same skill set?<p>Interesting to see how things like faster players change optimal play
dmos62将近 6 年前
Any chance for Google Research Rugby?
oliv3er将近 6 年前
&gt; The Football Engine is written in highly optimized C++ code, allowing it to be run on off-the-shelf machines, both with GPU and without GPU-based rendering enabled. This allows it to reach a performance of approximately 25 million steps per day on a single hexa-core machine.<p>Missed opportunity to use Rust for memory safety.
评论 #20136346 未加载
评论 #20136384 未加载