TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

OpenAI Universe

1021 pointsby samaover 8 years ago

29 comments

llSourcellover 8 years ago
Hey guys, it&#x27;s Siraj. OpenAI asked me to make a promotional video for it on my Youtube channel and I gladly said yes! You can check it out here:<p><a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=mGYU5t8MO7s" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=mGYU5t8MO7s</a>
评论 #13105365 未加载
评论 #13106525 未加载
评论 #13112918 未加载
评论 #13104150 未加载
评论 #13104426 未加载
评论 #13111949 未加载
mulcaheyover 8 years ago
With this platform (and Gym) it seems like a large part of their strategy for &quot;democratizing AI&quot; is to grow the amateur research community. By making it easier for an individual to play around and conduct experiments, they are hoping enable progress to emerge from anywhere instead of just from wealthy companies and elite universities.<p>It is also a great way to be able to track and organize what is being created rather than having to sort through amateur projects scattered across the web or research publications that often lack accompanying code.<p>Edit:<p>Some key ways they&#x27;re making it easier for amateurs:<p>* Starting point for problems to solve<p>* Way to get noticed (instead of needing a university&#x2F;company brand)<p>* Technological infrastructure for building and testing. The diversity of tools they brought together to build this platform is very impressive.
d_burfootover 8 years ago
Disclaimers: I cannot see the future. These are just my opinions. I really appreciate the work and money that SamA, Elon, and others have put into the OpenAI project. The Universe work in particular might help encourage young people, many of whom love video games, to study AI.<p>But I feel that contrarians, such as myself, have an ethical commitment to young people to voice our doubts and criticisms, so that they can avoid making a long journey down a career&#x2F;research path that leads to a dead end. That being said, I think this project leads in a very unpromising direction. Here are some reasons:<p>1. Games aren&#x27;t a good testbed for studying intelligence. In a game the main challenge is to map an input percept to an output action (am I drifting off the side of the road? Okay swerve right). The real challenge of intelligence is to find hidden abstractions and patterns in large quantities of mostly undifferentiated data (language, vision, and science all share this goal).<p>2. This platform is not going to help &quot;democratize&quot; AI. To succeed in one of these domains, contestants will need to use VAST amounts of computing power to simulate many games and to train their DL and&#x2F;or RL algos. DeepMind and others will sufficient CPU&#x2F;GPU power will almost certainly dominate in all of these settings.<p>3. Deep Learning, as it is practiced, isn&#x27;t intellectually deep. With a few exceptions, there is nothing comparable to the great discoveries of physics, not even anything comparable to the big ideas of previous AI work (A*, belief propagation, VC theory, MaxEnt, boosting, etc). Progress in DL mostly comes from architecture hacking: tweak the network setup, run the training algo, and see if we get a better result. The apparent success of DL doesn&#x27;t depend on any special scientific insight, but on the fact that DL algos can run on the GPU. That, combined with the fact that, except for the GPU, Moore&#x27;s Law broke down roughly 10 years ago, means that relative to everything else, DL looks amazingly successful - because all other approaches to AI are frozen in time in terms of computing power.
评论 #13106223 未加载
评论 #13106812 未加载
评论 #13106718 未加载
评论 #13108436 未加载
评论 #13112283 未加载
评论 #13108122 未加载
flaviojuvenalover 8 years ago
Related but slightly off-topic, there is a great sci-fi story by Ted Chiang (the same author who made the story behind Arrival film) about humans raising AIs in an artificial world. The premise is that if we want AIs to act like humans, we must teach them like we teach humans: <a href="http:&#x2F;&#x2F;subterraneanpress.com&#x2F;magazine&#x2F;fall_2010&#x2F;fiction_the_lifecycle_of_software_objects_by_ted_chiang&#x2F;" rel="nofollow">http:&#x2F;&#x2F;subterraneanpress.com&#x2F;magazine&#x2F;fall_2010&#x2F;fiction_the_...</a>
评论 #13107125 未加载
评论 #13108759 未加载
评论 #13108131 未加载
评论 #13109325 未加载
state_lessover 8 years ago
I&#x27;d love to see AI, using games, master the art of determining a depth for objects in the scene. If you ask a person, &quot;about how far away is that car?&quot;, they often give you an okay answer that is at least in the same magnitude as the actual distance 1 m, 10 m, 100 m, 1000 m. If AI could do that, you could then navigate an environment in the real world better using only a camera or two. So you start with a virtual world that looks real, train up the bot, then use it to navigate in the real world. Has this already been accomplished?
评论 #13106997 未加载
poppingtonicover 8 years ago
This is amazing! I was thinking of this problem when I saw a friend making a stop-motion video. The steps are super repetitive and I asked him, &quot;maybe an DeepMind Atari-style RL agent can learn how to do this?&quot; But I didn&#x27;t want to do what DeepMind did to emulate Atari games with an Adobe editing tool. This is an experiment that I can now run.
评论 #13104534 未加载
评论 #13104529 未加载
Cybioteover 8 years ago
This is astounding!<p>If requests are being taken, it would be useful to be able to search through the listed environments. And a poker environ for the internet section would be a good balance of fun, widely appreciable and a straight forward but very non-trivial environment.
评论 #13112929 未加载
NhanHover 8 years ago
This is a bit out there, but it would be fun if OpenAI can get one of the mega popular multiplayer games under this (WoW, League of Legends, DOTA etc.).<p>Imagine an AI team in League of Legends world championship!
评论 #13103928 未加载
评论 #13104037 未加载
评论 #13104078 未加载
评论 #13104252 未加载
评论 #13107514 未加载
shykesover 8 years ago
This is perhaps my favorite use of Docker ever.
评论 #13104247 未加载
jclayover 8 years ago
I noticed the OpenAI team wrote their own VNC driver in Go for performance reasons[0].<p>I would love to hear more about how they were able to achieve increased performance over other VNC drivers.<p>[0]<a href="https:&#x2F;&#x2F;github.com&#x2F;openai&#x2F;go-vncdriver" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;openai&#x2F;go-vncdriver</a>
评论 #13108776 未加载
soaredover 8 years ago
<a href="http:&#x2F;&#x2F;reddit.com&#x2F;r&#x2F;WatchMachinesLearn" rel="nofollow">http:&#x2F;&#x2F;reddit.com&#x2F;r&#x2F;WatchMachinesLearn</a> is about to get a lot more popular. I can&#x27;t wait. Also from the linked blog post, you can play with (against?) your agent in realtime:<p>&gt;You can keep your own VNC connection open, and watch the agent play, or even use the keyboard and mouse alongside the agent in a human&#x2F;agent co-op mode.
minimaxirover 8 years ago
Interesting announcement timing at 10:30 PM PST on a Sunday. :P<p>The list of third-party gaming partners is extremely impressive, and a Docker config helps resolve the dependency hell that some of the AI packages require.
评论 #13103801 未加载
Hydraulix989over 8 years ago
What is state of the art in reinforcement learning right now?<p><a href="https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;1602.01783" rel="nofollow">https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;1602.01783</a><p>Is there a way to deal with &quot;sparse&quot; training data (state, action, reward) triples -- sparse in &quot;state&quot;?
评论 #13105573 未加载
评论 #13105388 未加载
CodinMover 8 years ago
I&#x27;ll just go on a limb and consider this to be fucking awesome.
grondiluover 8 years ago
All the listed PC games environments are tagged as &quot;coming-soon&quot;<p><a href="https:&#x2F;&#x2F;universe.openai.com&#x2F;envs#pc_games" rel="nofollow">https:&#x2F;&#x2F;universe.openai.com&#x2F;envs#pc_games</a>
cingover 8 years ago
End game; I&#x27;d really like an AI agent for &quot;in real life&quot; tabletop games (like boardgames).
评论 #13103991 未加载
评论 #13106783 未加载
nooberminover 8 years ago
&gt;other applications<p>Any applications with a keyboard and mouse? Can I use emacs and have it start learning to code?
评论 #13105376 未加载
jakozaurover 8 years ago
Browser tasks seems to be a greenfield field with amazing potential.<p>What if AI can do anything what can human do you with a browser over the phone?<p>Also love &quot;bring your own Docker container format&quot;.
评论 #13104315 未加载
评论 #13104149 未加载
评论 #13104021 未加载
aratnoover 8 years ago
I just hope no self-driving vehicle is applying anything learned in GTA.
评论 #13103985 未加载
评论 #13103952 未加载
swahover 8 years ago
Layman question: isn&#x27;t adjusting &quot;hyperparameters&quot; similar to writing a algorithm for playing a game, using human intelligence?<p>Related to the blog post: <a href="https:&#x2F;&#x2F;openai.com&#x2F;blog&#x2F;universe&#x2F;" rel="nofollow">https:&#x2F;&#x2F;openai.com&#x2F;blog&#x2F;universe&#x2F;</a>
评论 #13105158 未加载
BaronSamediover 8 years ago
Unless I missed something it looks like the AI has to learn from screen pixels instead of getting game state data. I don&#x27;t like that approach at all. I understand that it&#x27;s easy to implement for OpenAI but I think having the game developers provide a real bot-capable API is much better. I hope the latter is what Blizzard will provide for their DeepMind collaboration.
评论 #13105856 未加载
评论 #13105703 未加载
评论 #13105711 未加载
评论 #13107092 未加载
thallukrishover 8 years ago
Being able to &quot;Infer&quot; from what it learns and &quot;Apply&quot; it to new scenarios in a general way is all about intelligence. I do not see how making it to win one game or one million will move it towards achieving general intelligence of this sort.
iotbover 8 years ago
Does OpenAI Universe communicate in any way with OpenAI remotely regarding activity in OpenAI Universe? Essentially, are there any call-home aspects to the code base? Or, is it possible to run this locally without any outside communication?<p>If there is remote communication, can you detail why and where it exists in code?
评论 #13109717 未加载
mariusz79over 8 years ago
I might be wrong but I think this was created mainly to monitor progress in AI research. If someone uses OpenAI Universe and can get better results than virtually everyone else, they will be able to get to them first.
评论 #13112218 未加载
NamTafover 8 years ago
From my initial reading, the end user can&#x27;t create environments? Is that a feature that I can expect will eventually come?
评论 #13105834 未加载
评论 #13105431 未加载
评论 #13105191 未加载
naveen99over 8 years ago
Too bad Iphone doesn&#x27;t support a vnc server. Would be nice to add some android apps if they could get permission.
mrfusionover 8 years ago
Is the users ai responsible for parsing the screen pixels that come back or does each game give you relevant events?
dcslackover 8 years ago
Some designers from Stripe absolutely helped with the design of this page.
daveloyallover 8 years ago
Didn&#x27;t we all agree to NOT let the AGI out of its box?<p>...That being said...<p>Instead of presenting the agent with a 2d plane of pixels, they should be presented with a sphere of pixels, with their POV inside.