News for hackers. Stuff that matters.
Re:OpenAI Team Update by fjdjdjnddn on Wednesday May 25, 2016 @06:42PM (#11773743)
So far I've seen lots of press releases and credential waving, but no actual development plan or progress. Until they start publishing papers a la Google, I'm putting this in the vaporware basket.
Re:OpenAI Team Update by gdb on Wednesday May 25, 2016 @07:56PM (#11774135)
Yep, I actually support this viewpoint. Judge us by our work, not by what we say about ourselves or whom we hire.

Fortunately, we'll have our first batch of papers coming out soon. (We also released OpenAI Gym a few weeks ago, but that's supporting infrastructure rather than novel research.) Hopefully we'll be able to earn our way out of the basket afterwards :).

Re:OpenAI Team Update by zodiac on Thursday May 26, 2016 @06:09AM (#11776808)
> we'll have our first batch of papers coming out soon

are you batching papers instead of releasing them when they're done?

Re:OpenAI Team Update by DanFeldman on Wednesday May 25, 2016 @06:47PM (#11773774)
True, but the people involved have much better things to do than join a vaporware research group. The people on this list alone (not to mention the heavily credentialed researchers from previous announcements) must have no problem finding meaningful work in this space, work that would produce results quite quickly. I'm inclined to believe there is something here and OpenAI is just ramping up.
Re:OpenAI Team Update by the_skeptic on Wednesday May 25, 2016 @07:26PM (#11773990)
Sure, you do have a point. That said the Meteor also had a stellar team, and so far I've been pretty disappointed with their output.
Re:OpenAI Team Update by jamesk_au on Wednesday May 25, 2016 @06:53PM (#11773813)
The public beta of OpenAI Gym, a toolkit for developing and comparing reinforcement learning algorithms, was released last month: https://openai.com/blog/openai-gym-beta/
Re:OpenAI Team Update by Analemma_ on Wednesday May 25, 2016 @08:03PM (#11774170)
This is kind of par for the course in non-profit AI research. OpenAI's spiritual parent MIRI also has something of a problem with lots of press releases and credential waving with few impactful papers or research.

That sounds a little harsh and I don't really mean in that way. Commercial AI/ML projects, since they have a clear goal they can measure against (play Go, drive a car, get better search results) tend to make meaningful progress that we can see, whereas the more philosophical and research-oriented non-profits don't have a yardstick like that, and probably won't until we have an inkling on how to actually make "strong AI". So this organization will probably seem like vaporware for some time, but hopefully it will prove its worth in the long run.

Re:OpenAI Team Update by gdb on Wednesday May 25, 2016 @08:13PM (#11774204)
Don't worry, we can and will do better than that. Please hold us to a higher standard!
Re:OpenAI Team Update by foobarqux on Wednesday May 25, 2016 @09:20PM (#11774530)
How does the group work exactly? Is everyone working on their own thing? Is the group focused on a specific goal?
Re:OpenAI Team Update by xamdam on Wednesday May 25, 2016 @09:58PM (#11774690)
a) Not sure if MIRI is a spiritual parent of OpenAI. It's a different approach to the problem, and I'm not sure MIRI would spawn OpenAI if they had the resources.

b) Also, unfair to MIRI. They are not only researching the area but also creating it. A lot of early work want into socializing AI risk as an idea, and in that sense pretty much every future effort in this area owes to them. I've given noticeable dollars to MIRI and pretty satisfied with ROI (of course always room for improvement), would do it again.

Re:OpenAI Team Update by AndrewKemendo on Wednesday May 25, 2016 @08:23PM (#11774248)
In my opinion it's impossible for them to do anything of substance - unless they turn into a lobbying body to drive legislative regulation of AI/AGI development.

Even with concrete plans/objectives ML/DL/AI projects rarely live up to their hype. IMO An AGI project that isn't focused around observable learning models (eg, not around "consciousness" etc...) isn't going to make much progress.