Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

OpenAI bots competing against Humans right now

twitch.tv

Post reply on HN
Print view View original

91–100 of 138 posts

Re: OpenAI bots competing against Humans right now

#91

Post by halflings »

fizx wrote: post #72
halflings wrote: post #35

Earlier quoted context omitted.

> in starcraft, most "strategies" are heavily scripted build orders that almost never deviate from a handful of openings (similar to opening moves in chess) and a large amount of winning comes from the ability to quickly give orders to your troops ("micro") As somebody who plays StarCraft casually (gold/low plat in ladder), this is not true. It's even less true for pro players. The level of strategy in StarCraft is i…

I wonder how far you'd get with a bot that macros perfectly but also A-moves 2-3 groups.

What does "macro perfectly" mean? If it does the same strategy over and over, you just scout, find its strategy, and go for the counter. Its macro will be useless if it has the wrong type of unit.

In a way, the built-in AIs "macro perfectly", but they are terrible at strategy and fighting (because even fights are not just a matter of gimmicks, you need to split units in a special way, send diversions, attack at the same time from multiple fronts, etc.)

Re: OpenAI bots competing against Humans right now

#92

Post by red75prime »

ethbro wrote: post #77
asdfasgasdgasdg wrote: post #71

Earlier quoted context omitted.

Whether the AI teaches us or not, once it surpasses us, we will not catch up. This is the case for every game AI that has ever surpassed human performance so far, and there's no reason to expect that this will be different in the case of DotA.

> game AI that has ever surpassed human performance so far Am I missing something, or does that set consist of Checkers, Chess, and Go so far? (presumably with analogous misc games of comparable complexity) Discounting the reaction time wins, I'd say the sample size is too limited to generalize to eventual AI behavior in more complex / open-ended games. Extrapolation was the cause of the last AI winter.

Statistics is not everything. Which hypothesis can lead to the outcome "AIs will not be able to beat humans in every game"?

I can see following hypotheses (in no particular order):

1. Human brain is the optimal solution in the space of all computational devices capable of playing games, and we can only approach it.

2. To do computation human brain employs some physical processes, we will not be able to replicate in the foreseeable future.

3. Human brain do not produce general intelligence, so we will not be able to replicate it as such a task is outside of the scope of our limited intelligence (while playing games isn't).

4. Human brain uses metaphysical abilities to do cognition, we will not be to replicate them at all.

5. Human brain is a local optimum, but the space of potential AIs' constructions is too huge to explore in the lifetime of our civilization, so we will be stuck at this local optimum with marginal improvements.

I don't see any of them as sufficiently likely, but your mileage may vary.

Re: OpenAI bots competing against Humans right now

#93

Post by badpun »

nopinsight wrote: post #82

What the cases so far tell us: Once an AI beats humans who have spent 10,000 hours practicing a skill, it is a matter of time before it beats the best professionals in the field. Cases it already happened: board games such as Chess and Go, Poker, diagnostics of certain diseases using medical images Cases where AI is still clearly inferior: video understanding, natural language understanding , motor control esp of han…

> Poker

This is a misnomer.

The only variant of Poker where AI beats humans is heads-up (two-player) variant, which is simplest form of poker and also rarely played. The AI was (marginally) beating humans there by playing a game-theory-optimal strategy. For poker games with 3+ players, the GTO strategy (Nash equilibrium) no longer exists, so AIs need to use more standard techniques (search-based, reinforcement learning etc.), which are, at the current state of the art, laughably weak at poker.

Not to mention, that in poker the actual hierarchy of players' skill is not 100% obvious. You could distinguish at least two areas of skills:

- play vs other experts

- play vs amateurs/weaker players. Here' the goal is not to come out ahead (which, in long term, is a given), but to _maximize_ the dollar amount taken from these players, which is a skill in itself.

Re: OpenAI bots competing against Humans right now

#94

Post by skgoa »

nopinsight wrote: post #82

What the cases so far tell us: Once an AI beats humans who have spent 10,000 hours practicing a skill, it is a matter of time before it beats the best professionals in the field. Cases it already happened: board games such as Chess and Go, Poker, diagnostics of certain diseases using medical images Cases where AI is still clearly inferior: video understanding, natural language understanding , motor control esp of han…

OpenAI is far from beating "the best professionals" of dota in a full dota match, though.

Re: OpenAI bots competing against Humans right now

#95

Post by skgoa »

kmnc wrote: post #34

Very impressed with the bots. The main strategy I see from the bots that is different from pro meta is how they spam abilities aggressively. Now the main limitation is 5 unkillable couriers which enable this. Watching, I do feel like given more practice humans would beat this version of the ai easily. It looks exploitable, and still having many rules limitations it has a ways to go. It will win in the end though, tha…

The bots also have perfect knowledge of enemy hp/mana and the relative positioning through the API. In fact the bots are gifted perfect micromanagement through the API. This enables them to do things most humeans wouldn't try, because the risk of messing up and giving the enemy a huge advantage is too high.

Re: OpenAI bots competing against Humans right now

#96

Post by skgoa »

21 wrote: post #49
hackandtrip wrote: post #37

Earlier quoted context omitted.

It would be really interesting to see what bots will do once humans will try to exploit this strategy, like adapting to the lane movement they always do after winning a lane etc, humans did that once today but it may be used better. Excited for the TI to see less limitation, a better shop system (lot of money throw away) and maybe better ward.

The next AI will always be much better (see AlphaGo progression). Humans, not so much, as in all top-level competitions, human abilities improve minimally at the top, because we have millions of humans competing against each other until the plateau of human performance is reached. Then you can push that a bit more with drugs (see doping in sports). And after that, you are pretty much done. So it's only a matter of ti…

> Humans, not so much, as in all top-level competitions, human abilities improve minimally at the top

That's just not true in doto. The player base improves quite a bit over time. The top pro plays from only a few years ago are not impressive anymore.

Re: OpenAI bots competing against Humans right now

#97

Post by skgoa »

hackandtrip wrote: post #30
TaupeRanger wrote: post #21

Earlier quoted context omitted.

It's not that exciting and it's very easy to beat humans at games when your reaction times are far faster and you have access to more information at a single moment in time. It's like being impressed by a calculator.

Reaction times are limited to 200ms; it's all about the meta and the group tactics, and bot are apparently really good at it.

The bots have perfect information (when the enemy is in vision) and perfect micromanagement through the API. A huge part of the challenge in dota is processing the viuals and audio cues from the game quickly and acting the limited information you have precisely. The bots get to take a shortcut around all of that.

Re: OpenAI bots competing against Humans right now

#98

Post by skgoa »

lawrenceyan wrote: post #46

Expecting 3-0 slam dunk here. Consistently surprised at the ability of tried and true basic reinforcement learning at completing challenging tasks. I totally expected some kind of neccessary breakthrough in the RL field before a real time game like Dota 2 could be beaten.

This still isn't a real game of doto, tho

Re: OpenAI bots competing against Humans right now

#99

Post by nopinsight »

skgoa wrote: post #94
nopinsight wrote: post #82

What the cases so far tell us: Once an AI beats humans who have spent 10,000 hours practicing a skill, it is a matter of time before it beats the best professionals in the field. Cases it already happened: board games such as Chess and Go, Poker, diagnostics of certain diseases using medical images Cases where AI is still clearly inferior: video understanding, natural language understanding , motor control esp of han…

OpenAI is far from beating "the best professionals" of dota in a full dota match, though.

Yes. My prediction from the observation above, which agrees with predictions from many others, is that it is a matter of time before that happens (likely within 1-2 years if they continue to work on it).

Re: OpenAI bots competing against Humans right now

#100

Post by Jabbles »

NegatioN wrote: post #45

My thoughts on the AI's performance so far knowing a bit about both DOTA and machine learning: - Item-usage for things such as smoke and wards (which were recently added to their reportoire) are not well captured by the bots yet. And the buying of wards were confirmed as a scripted event. It seems hard for them to capture the long term sparse reward of these. Smoke might not be needed by a perfect agent, but wards sh…

Otherwise it would require all 5 agents gathering there, expending their magical abilities and investing a lot before actually seeing a reward. (which is unlikely to happen)

This is the key - if we had machine learning techniques that allowed it to reason at the level of "what happens if I kill this?", we could explore more of the interesting state space more quickly. Perhaps there are advances in intrinsic reward systems that allow this.

Post reply on HN