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。

Mastering Atari Games with Natural Intelligence

31 点作者 alex_c4 个月前

5 条评论

vrighter4 个月前
Something that can excel at "mastering atari games with natural intelligence" is called a speedrunner. It is not made of silicon.
internet_points4 个月前
doesn't "natural intelligence" normally mean human intelligence?
评论 #42832112 未加载
SonOfLilit4 个月前
This smelled like BS the moment they invoked Karl Friston, but when I noticed it&#x27;s a publicly traded company (without any available products or research artifacts) I became _extremely_ suspicious.<p>The only non-scammy thing I could find about them is that Friston seems to work there.
评论 #42829390 未加载
评论 #42829112 未加载
djmips4 个月前
Very nervy trademarking genius and calling this natural intelligence.
low_tech_love4 个月前
“If, metaphorically, the methods used by DQN and Agent57 are gas-guzzling Hummers and those used to tackle the Atari 100k challenge are like a fuel-efficient Prius, then our approach used on Atari 10k is like a Tesla, a hyper-efficient alternative architecture.”<p>Eh, what?
评论 #42829406 未加载