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。

Generative Flow Networks

168 点作者 selimonder大约 3 年前

6 条评论

cs702大约 3 年前
For a quick, easy-to-digest introduction to the topic, there&#x27;s a fantastic video with explanations and an interview with Bengio that uses Galton peg boards as a starting point for understanding generative flow networks. Highly recommended if you&#x27;re a researcher or practitioner interested in learning about this:<p><a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=M49TMqK5uCE" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=M49TMqK5uCE</a>
manux大约 3 年前
Hi, first author of the original NeurIPS paper here! Someone just shared this HN post with me. I&#x27;ll go through the comments but happy to try to answer questions as well.
评论 #30593120 未加载
评论 #30592668 未加载
soraki_soladead大约 3 年前
Here&#x27;s the referenced tutorial: <a href="https:&#x2F;&#x2F;milayb.notion.site&#x2F;GFlowNet-Tutorial-919dcf0a0f0c4e978916a2f509938b00" rel="nofollow">https:&#x2F;&#x2F;milayb.notion.site&#x2F;GFlowNet-Tutorial-919dcf0a0f0c4e9...</a><p>(Some of the links appear to require signing in to Notion but this one doesn&#x27;t.)
mark_l_watson大约 3 年前
I started reading through the linked tutorial, but I think I need to first back up and review reinforcement learning (I took 2 Coursera classes in RL, and helped with one project at a previous job, so I am at a beginner stage).<p>I love deep learning, a practical way to solve many problems (and most of my work over the last 6 years) but DL seems incomplete as far as explainability, causation, etc. Anyway, I just retired last week (mostly to take care of my wife who is having a health crisis) so I have plenty of time to study GFN, probably starting with re-taking the 2 Coursera classes on RL and then finishing the RL specialization.
awelkie大约 3 年前
Having just skimmed the linked tutorial, it looks like GFlowNets could be useful for automated theorem proving. Anyone with more knowledge know if this is possible?
评论 #30591938 未加载
评论 #30590628 未加载
ramesh31大约 3 年前
Correct me if I&#x27;m wrong, but it seems like SOTA for AI now is entirely in developing more and more complex sets of convolutions for deep learning?
评论 #30591686 未加载