They themselves said it was trained on NVIDIA chips, so I’m not sure where you got that it wasn’t. It was trained on the less capable versions sold for the Chinese market.
I see, thank you for pointing that out. Then I’d rephrase, how do we know for sure that it wasn’t trained on the most advanced Nvidia chips? Did anyone outside of China replicated the training?
Why do you think there's a longer interval between inspections for Teslas? The interval for Hauptuntersuchung in Germany (which this article is about) is 2 years (except for the first one which happens after 3 years), regardless of car brand.
And a windshield wiper needing replacement wouldn't count as a "significant fault" that they force you to come back for before getting certified.
And my claim is that during a yearly maintenance any faults are found and corrected before the HU inspection. If you don't have a yearly maintenance then you have a longer time period for faults to develop. Or for items to expire, like a first aid kit.
Not German, but I do wonder if it's common to schedule your oil change and other maintenance just before your state inspection. I would certainly do that; get the dealer to fix all pending issues and get the car into tip-top shape just before a mandatory inspection. If this is the case, you would expect more problems with Tesla, as there isn't a dealer mechanic minimizing issues just before inspection.
It’s not uncommon to ask a garage to take care of the whole TÜV inspection process - they’ll do a pre-inspection, correct any issues they find, take the car to the TÜV, and then also fix any issues in the (unlikely) event that it fails.
This likely wouldn’t be the process with a Tesla, given there are no brick-and-mortar Tesla mechanics, and it’s less likely (given the absence of servicing) you’d have an ongoing relationship with a traditional garage for the car.
Unfortunately, some companies' HR departments actively encourage "trivia questions", because they ask interviewers to ask every candidate the same questions so that a "fair" comparison between candidates.
I think you've touched on a crucial point here about making the comparison "fair". In my view, I find the "trivia questions" approach to be an abdication of responsibility for the difficult, nebulous task of assessing a candidate's abilities. It gives you an objective measure by which to compare candidates... even though the measure itself is usually arbitrary and irrelevant.
I think by 'seal the doom' he just means that this result shows that OpenAI is almost definitely going to be able to defeat a pro team in an unrestricted game of DotA.
Which I'm still not completely sold on. It's likely, but the remaining restrictions aren't trivial by any means. There's an outside chance that removing one or more of them is going to brickwall their progress.
One should keep in mind that some of the restrictions were in place to prevent the bots from having too easy of a time. For example, the anti micro/illusion rule was intended to limit the obviously superior micro coordination of the bots.
I'm not sure that's true? I can see the bots being utterly terrifying with meepo in a teamfight - but would need supports stacking, proper farm prioritisation (much more use of jungling and ancients), etc etc.
I genuinely believe the bot would win a game of turbo against any team in the world. But remove _all_ of the restrictions and it's not clear that it doesn't just lose at the moment
They specifically said they would have to implement a special case for heroes that control more than one unit in the future.
So you're saying that even before they set up a rule about microing illusions to protect humans from a feature that they have not yet implemented nor, I assume, have trained the model on?
Not only that, but also lets not forget humans learn as well. Meaning the more games players play against the bot the better they would become at understanding and defeating it.
> Why wouldn't it have good coordination? A bot has access to a perfect model of how the other bot would act - itself.
As far as I know it is five (Hence the name) individual AI instances controlling each character and with basically no AI to AI communication.
It is not one overriding AI controlling all five.
I have no idea if the AI instance controlling each character is identical though, if so then your statement still holds true I guess (Assuming each AI has the exact same information to work with which might be the case). It would be interesting to see if AIs specialised.
There's a presiding team value function that impacts and steers team play. The bots 'communicate' through this. There's nothing magical going on.
As a counter bot strategy, I'd work on how to break and trick it using multiple-stepped logic that an optimization function would be unable to see beyond. I'd also use varying tactics of chaotic/sporadic configurations. The bot isn't 'playing fair' nor should a human w/ intelligence. The advantage being that a human can think along a multitude of strategies and adapt. The bot is only optimizing some steps ahead.
Their 1v1 bot was defeated in this manner and it just goes to show what true intellect and superiority is. I've played random pub games w/ little to no communication and have had all other 4 players converge on different strategies based on a perception of what's going on. If someone decided to cheese/snowball, you simply wait it out and let them push themselves into a nightmare. I saw little to none of this in the games I watched which leads me to question the intelligence of said 'pros'.
The team value function is just a hyperparameter that describes how greedy the individual agents are. At the start of training the team spirit is 0 and the bots are only rewarded for their own actions. This encourages them to learn basic micro skills, like last hitting. As training progresses the team spirit is increased. When it finally reaches 1, the bots value a reward for a teammate as highly as a reward for themselves.
The actual source of the "communication" is not the team spirit parameter, but the basic fact that the bots have been trained together and they receive the same inputs when making decisions. Unlike humans, who have a limited focus to their attention, the bots can look at the whole map at once. They don't need to communicate because the already "know" what their allies will do when given the same input.
Physics and other disciplines were broadly using arxiv before ML did it. When I switched from physics to ML (CS) I was surprised that arxiv isn't used more.
Depends on what your industry is. If you're compiling c++ for mobile, clang is pretty much your only choice for iOS and the default choice for Android ...
It has been the default compiler in base on FreeBSD since version 10, and FreeBSD is now built with it. There was also a push back around 2014 or so to get much of the ports tree built with it, too.
DragonFly BSD has been buildable with clang since around 2014, albeit using clang from packages/ports. In 2017 (release 50) the DragonFly people started work on pulling it into base.
OpenBSD switched building on x86/amd64 to clang in 2017 and building on armv7 to clang in 2018.
NetBSD has included clang in base but as far as I know does not (yet) build with it by default on any architecture.
I only mentioned Linux because I knew that my argument wouldn't work for BSDs! But in my defense the situation, at least originally, was more political than technical (at least as far as FreeBSD was concerned, I don't follow the other BSDs too closely). GCC switching to the GPLv3 was a big no-no for many people in the BSD world. As a result they were at the forefront to port the system to clang since they were stuck with an obsolete GPLv2 GCC version.
> Because our training system Rapid is very general, we were able to teach OpenAI Five many complex skills since June simply by integrating new features and randomizations. Many people pointed out that wards and Roshan were particularly important to include — and now we’ve done so. We’ve also increased the hero pool to 18 heroes. Many commenters thought these improvements would take another year.
The linked commenters thought that getting to "real dota" (more than 100 heroes, captains mode instead of random, ...) would take another year. So I don't think it's fair to make that statement.
Edit: Don't get me wrong, I think the improvements are very nice, but pointing to people saying "these people thought we would need a year, we did it in under a month!" is not what you should do if you didn't actually do what the linked people stated.
It seems like the 5 invulnerable couriers restriction is something that will have a huge influence on how the early game is played and is something that the humans won't have any experience taking advantage of.
