Comment by Strom | Hacker News Reader

Comment by Strom | Hacker News Reader

> Agents were capped at a max of 22 agent actions per 5 seconds, where one agent action corresponds to a selection, an ability and a target unit or point, which counts as up to 3 actions towards the in-game APM counter. Moving the camera also counts as an agent action, despite not being counted towards APM.

I'm happy to see that they've greatly improved the APM cap. During the earlier showmatch they had an extremely naive cap on the average APM during the whole game. Which meant that the AI could take it easy for most of the time and then during battle sequences hit peaks of 2000 APM. Also the previous cap was based on the in-game APM counter, which doesn't count some things like moving the camera, which is now addressed. The current state sounds a lot better.

However it seems still superhuman in mechanical non-strategy ways, e.g. a human can misclick (click too early/late when moving the mouse, or just miss the target with the cursor completely) or do accidental double clicks etc. These end up being very costly mistakes against a zero mechanical mistake AI. Which in turn means that the AI can win with an inferior strategy due to the extra gains it has via superior mechanical accuracy. In other words this artificial intelligence is still under significacnt artificial motor skills welfare, and thus even if it starts beating pro players we shouldn't be too quick to talk about how it's the intelligence part that got it the win.

All of that said, I'm liking the progress and am excited to see what they achieve with this next. Would love another showmatch against pro players.

Yes, those might have some impact, but it is clear that the progress is there, with this new APM cap and camera movement etc.

You can also see in replays that the AI often makes mechanical mistakes, missing spells, missing units, even ordering wrong units from outside the screen - so it surely seems that if it's win rate was conditioned in any strong way on its sheer mechanical ability, it would have learned to not make any such mistakes. Since the mistakes are clearly there - it seems that its power comes primarily from somewhere else, likely from the AI ability to choose the right actions, not from mechanical power of executing them perfectly.

Thus it is not likely that "this artificial intelligence is still under significacnt artificial motor skills welfare". They have consulted top players before they set this AI up, and if those details were important, they would have baked them into the limitations list already.

The mistakes it makes are due to bad decisions. There have been no claims by DeepMind that they have some sort of chaos engineering [1] going on where the AI decides one thing and then the output system actually does another thing.

Also I think you overestimate the AI/IT knowledge of these top players that they're consulting. I have great respect towards them, but they're not renaissance men [2] who both play 10 hours of StarCraft per day and also know the subtleties of how computers work, not to mention bleeding edge AI. You can watch the previous showmatch [3] to see how DeepMind people lack knowledge of StarCraft and how the pro players they're consulting lack understanding of the AI and both are learning new things live on air. Its obvious that their cooperation is bearing fruit as they spend more time talking to each other, as evidenced by the new APM limits. However I'm willing to bet that they would reach many more good conclusions if they just continue working together.

--

[1] https://en.wikipedia.org/wiki/Chaos_engineering

[2] The problem with both parties (the pros & deepmind) is that they're so overspecialized. I'm nowhere near as good as them at their respective fields, but I am a professional programmer and diamond in StarCraft II. In addition I've built StarCraft II AI myself, although with different goals related to finding optimal strategies.

[3] https://www.twitch.tv/videos/369062832

You're not giving them enough credit. Oriol Vinyals (head of this project) is both one of the leading AI researchers + was highest ranked StarCraft player in Spain back in high school.

Certainly the game is different, meta has changed, etc. But he's definitely not an amateur and is probably in the top percentile of players.

>[2] The problem with both parties (the pros & deepmind) is that they're so overspecialized. I'm nowhere near as good as them at their respective fields, but I am a professional programmer and diamond in StarCraft II. In addition I've built StarCraft II AI myself, although with different goals related to finding optimal strategies.

So many things wrong with this comment.

You're nowhere near a professional StarCraft player if you're in Diamond league (I play casually and I'm on high plat, bordering Diamond), and Oriol Vinyals, the lead research scientist behind this project, is one of the most renowned scientists in the field and used to play StarCraft at a professional level. They also said that other employees at DeepMind are at Masters level, and helped test the AlphaStar.

You quoted me saying that I'm not as good as them and then you say that my statement is wrong because I'm not as good as them. We seem to be in agreement. Maybe you misread? Anyway that was just a footnote to point towards my generalist nature, because I think that's a fundamental reason why I immediately see these things while for them it takes time to realize. Also to be even more fair towards them, my generalist knowledge around computer controlled gaming goes way beyond StarCraft II as I've written bots for many different games.

My main argument revolves around APM. Oriol Vinyals might be great, but I've also seen him make a statement on video in 2019 [1] how restricting the AI based on average APM during the whole game is reasonable. He has blindspots that someone like me can immediately spot.

--

[1] https://www.twitch.tv/videos/369062832

These live on air admissions of prior ignorance that you mention maybe for illustrative purposes, a PR gag to explain the difficult approaches to each other they had taken in development. It's like late night shows.

You must be one of the people who think really tv programs were really real.

Why don’t they ever just have a virtual mouse api or whatever, with some lag and some jitter just to make it an actual apples to apples comparison?

The AI is supposed to call this API. Then actions per minute would be irrelevant.

I'm just guessing but I think it would be easier to start from a perfect system and progressively work towards an accurate imperfect system rather than the other way of guessing an imperfect system and working back towards an accurate imperfect system.

Easier but that’s cheating LMAO

Grandmaster level players make those kinds of mistakes extremely rarely and if they’re the only difference between winning and losing, the ai is an extraordinary achievement.

It's not exactly the same as a showmatch, but you can find lots of YouTube videos analyzing replays of AlphaStar playing on the ladder. For example, this game against 2018 world champion Serral: https://www.youtube.com/watch?v=_BOp10v8kuM

And even with limited APM, the response time can still be superhuman. Bit disappointed that they didn't add some human-level latency to the agents' actions.

How much do misclicks or accidental double clicks matter at the pro level?

A lot. In the same way that errors reduce data bandwidth, misclicks reduce effective APM in hectic situations. Classic examples are moving marines into banelings or units into disruptor shots.

Sorry, I asked that wrong. Clearly misclicks matter at the pro level, as pros are going to exploit their competitors mistakes ruthlessly. What I should have asked is, at the pro level, how often do misclicks affect the outcome of a match?

Every game of Starcraft is decided by an extremely large amount of small events. It is very rare that a single event or misclick is responsible for the outcome of a game, but it happens occasionally.

Since players take so many actions during the course of the game (well into the tens of thousands), inevitably some clicks will be sub-optimal, and they all have a tiny impact on the outcome. Some professional players do specific exercises to improve their clicking accuracy in order to gain efficiency by reducing misclicks, but generally clicking accuracy is not considered a big factor compared to raw speed. Most players try to attain the highest possible clicking speed while maintaining an accuracy level that is "good enough".

from what I've heard from a starcraft player, what you've just described is just how it is. You don't need good strategies if you have superior techniques (apm & precision). It's not really a game of intelligence (after you've got the basics down).

an SC robot with alphastar brain--learning to use a mouse for a start--would probably only shift the blame to those super human materials. They should put alpha* into a human! That's where it's going anyway, isn't it?

Reply on news.ycombinator.com