• Log InLog In
  • Register
Liquid`
Team Liquid Liquipedia
EDT 00:26
CEST 06:26
KST 13:26
  • Home
  • Forum
  • Calendar
  • Streams
  • Liquipedia
  • Features
  • Store
  • EPT
  • TL+
  • StarCraft 2
  • Brood War
  • Smash
  • Heroes
  • Counter-Strike
  • Overwatch
  • Liquibet
  • Fantasy StarCraft
  • TLPD
  • StarCraft 2
  • Brood War
  • Blogs
Forum Sidebar
Events/Features
News
Featured News
RSL Season 1 - Final Week6[ASL19] Finals Recap: Standing Tall12HomeStory Cup 27 - Info & Preview18Classic wins Code S Season 2 (2025)16Code S RO4 & Finals Preview: herO, Rogue, Classic, GuMiho0
Community News
Esports World Cup 2025 - Brackets Revealed13Weekly Cups (July 7-13): Classic continues to roll4Team TLMC #5 - Submission extension3Firefly given lifetime ban by ESIC following match-fixing investigation17$25,000 Streamerzone StarCraft Pro Series announced7
StarCraft 2
General
Who will win EWC 2025? The GOAT ranking of GOAT rankings Esports World Cup 2025 - Brackets Revealed RSL Revival patreon money discussion thread Weekly Cups (July 7-13): Classic continues to roll
Tourneys
FEL Cracov 2025 (July 27) - $8000 live event Sea Duckling Open (Global, Bronze-Diamond) RSL: Revival, a new crowdfunded tournament series $5,100+ SEL Season 2 Championship (SC: Evo) WardiTV Mondays
Strategy
How did i lose this ZvP, whats the proper response Simple Questions Simple Answers
Custom Maps
External Content
Mutation # 482 Wheel of Misfortune Mutation # 481 Fear and Lava Mutation # 480 Moths to the Flame Mutation # 479 Worn Out Welcome
Brood War
General
Flash Announces (and Retracts) Hiatus From ASL ASL20 Preliminary Maps BW General Discussion BGH Auto Balance -> http://bghmmr.eu/ Starcraft in widescreen
Tourneys
Cosmonarchy Pro Showmatches [Megathread] Daily Proleagues CSL Xiamen International Invitational [BSL20] Non-Korean Championship 4x BSL + 4x China
Strategy
Simple Questions, Simple Answers I am doing this better than progamers do.
Other Games
General Games
Stormgate/Frost Giant Megathread Path of Exile Nintendo Switch Thread CCLP - Command & Conquer League Project The PlayStation 5
Dota 2
Official 'what is Dota anymore' discussion
League of Legends
Heroes of the Storm
Simple Questions, Simple Answers Heroes of the Storm 2.0
Hearthstone
Heroes of StarCraft mini-set
TL Mafia
TL Mafia Community Thread Vanilla Mini Mafia
Community
General
US Politics Mega-thread Things Aren’t Peaceful in Palestine Russo-Ukrainian War Thread Stop Killing Games - European Citizens Initiative Summer Games Done Quick 2025!
Fan Clubs
SKT1 Classic Fan Club! Maru Fan Club
Media & Entertainment
[Manga] One Piece Movie Discussion! Anime Discussion Thread [\m/] Heavy Metal Thread
Sports
Formula 1 Discussion TeamLiquid Health and Fitness Initiative For 2023 2024 - 2025 Football Thread NBA General Discussion NHL Playoffs 2024
World Cup 2022
Tech Support
Computer Build, Upgrade & Buying Resource Thread
TL Community
The Automated Ban List
Blogs
Men Take Risks, Women Win Ga…
TrAiDoS
momentary artworks from des…
tankgirl
from making sc maps to makin…
Husyelt
StarCraft improvement
iopq
Trip to the Zoo
micronesia
Customize Sidebar...

Website Feedback

Closed Threads



Active: 744 users

StarCraft II: DeepMind Demonstration: Jan 24 - Page 29

Forum Index > SC2 General
585 CommentsPost a Reply
Prev 1 27 28 29 30 Next All
figq
Profile Blog Joined May 2010
12519 Posts
Last Edited: 2019-01-27 12:29:45
January 27 2019 12:25 GMT
#561
To remind you that we've seen various different agents who trained together, not one entity. Some of them focused on blink stalkers and won by inhumanly good micro. Those seemed like easy victories. But really we've seen a great variety of approaches - aggressive, defensive, tactical and strategic.

The key here is that the AI trained with itself - AI vs AI in ~200 human years of training. That doesn't give it realistic perspective on what works best vs humans. If the AI was let to compete on ladder vs real humans, I bet it would focus exclusively on things like blink stalkers all the time - not just in some random instances - that give it inherent mechanical advantage no human can match.
If you stand next to my head, you can hear the ocean. - Day[9]
TaKeTV
Profile Blog Joined December 2010
Germany1199 Posts
January 27 2019 12:46 GMT
#562
On January 27 2019 21:25 figq wrote:
To remind you that we've seen various different agents who trained together, not one entity. Some of them focused on blink stalkers and won by inhumanly good micro. Those seemed like easy victories. But really we've seen a great variety of approaches - aggressive, defensive, tactical and strategic.

The key here is that the AI trained with itself - AI vs AI in ~200 human years of training. That doesn't give it realistic perspective on what works best vs humans. If the AI was let to compete on ladder vs real humans, I bet it would focus exclusively on things like blink stalkers all the time - not just in some random instances - that give it inherent mechanical advantage no human can match.


On the one hand it obviously would be smart and the logical thing to do: exploit your own strength and abilities - aka unmatched mechanics. On the other hand that really wouldn't give a realistic approach of an Ai actually playing against a human. Humans need to physically move the mouse against friction, the distance etc. Also observation and reaction play a role as well. I find it interesting but what we saw is far away from actually an Ai playing / outplaying humans but rather a mechanically superior agent. And the biggest task will be learning to play the game with no knowledge of the map, actually having to fight for information and telling the difference between a fake, a human mistake etc.
Commentator
Polypoetes
Profile Joined January 2019
20 Posts
January 27 2019 13:13 GMT
#563
On January 27 2019 21:25 figq wrote:
If the AI was let to compete on ladder vs real humans, I bet it would focus exclusively on things like blink stalkers all the time - not just in some random instances - that give it inherent mechanical advantage no human can match.


But it is not really possible to train the AI on the ladder. How do you get 200 years of game time worth of your bot playing vs humans on the ladder? Adjusting the weights of the NN based on how well it does vs humans is just really tricky because you will have a much smaller dataset. And the dataset could have all kinds of biases. And that even ignores the fact that people could be trolling. We have seen chat NNs say racist things because of people trolling.

Secondly, even if you can train a NN to exploit mistakes humans make, maybe that is actually an inferior way to play. The NN will make deliberate mistakes 'knowing' that humans cannot exploit it. But another bot will. So you will just be muddying the way the NN plays with weakness.

And third, things the NN does that seem like a mistake, once it gets so strong it beats humans, it is not clear we humans can recognize a mistake. Our criticism of what it does is only justified if we can show that we can exploit it as a weakness. Humans cut corners too. Maybe it knows it can cut corners. If the AI builds 7 observers vs Mana, maybe it does that because it is minimizing the risk to lose. And it judged the risk of losing by being outmacroed many magnitudes lower than being killed by DTs. So it just overproduced observers and then puts them all near it's biggest army.

So while it seems like a mistake, and it probably is, humans stepping in and manually adjusting weights doesn't work because you have a very poor idea of what you are doing.
Haukinger
Profile Joined June 2012
Germany131 Posts
January 27 2019 14:14 GMT
#564
It's funny how people deem a korean terran exploiting marines with superior mechanical skill the highest level of play but if an AI uses blink stalkers it's just stupid with high apm.
snakeeyez
Profile Joined May 2011
United States1231 Posts
Last Edited: 2019-01-27 18:14:12
January 27 2019 18:10 GMT
#565
On January 27 2019 21:46 TaKeTV wrote:
Show nested quote +
On January 27 2019 21:25 figq wrote:
To remind you that we've seen various different agents who trained together, not one entity. Some of them focused on blink stalkers and won by inhumanly good micro. Those seemed like easy victories. But really we've seen a great variety of approaches - aggressive, defensive, tactical and strategic.

The key here is that the AI trained with itself - AI vs AI in ~200 human years of training. That doesn't give it realistic perspective on what works best vs humans. If the AI was let to compete on ladder vs real humans, I bet it would focus exclusively on things like blink stalkers all the time - not just in some random instances - that give it inherent mechanical advantage no human can match.


On the one hand it obviously would be smart and the logical thing to do: exploit your own strength and abilities - aka unmatched mechanics. On the other hand that really wouldn't give a realistic approach of an Ai actually playing against a human. Humans need to physically move the mouse against friction, the distance etc. Also observation and reaction play a role as well. I find it interesting but what we saw is far away from actually an Ai playing / outplaying humans but rather a mechanically superior agent. And the biggest task will be learning to play the game with no knowledge of the map, actually having to fight for information and telling the difference between a fake, a human mistake etc.


I feel like it already showed some of this. It could see the entire map unlike a human, but it never had any different vision. It had to fight for and scout all of its vision just a person in all the games. It just didnt need the camera which is a big advantage.
I think they still have a long ways to go though they need to play more matches with a limited camera, and there is a large variety of strats in this compared to chess. When it plays the best players its going to have a tough time if its limited in its micro to be more like a human.
That is the point of this though overcoming these challenges. If it won with this demonstration easily then it was not much of a challenge to begin with. The fact it still has a ways to go is why they choose to try to win at starcraft 2. its probably one of the hardest games left to beat. The other was GO and they beat that which was amazing.
Jasper_Ty
Profile Joined July 2017
101 Posts
Last Edited: 2019-01-27 19:42:31
January 27 2019 19:41 GMT
#566
I think people are forgetting just how impressive having a bot with superb, dynamic micro and great tactics is. A lot of 'perfect micro' bots are incredibly exploitable. For example, I could naively make a marine split bot that makes marines kite against banelings forever, but the moment an engagement happens on creep it just falls apart because speed banes are faster than stimmed marines on creep. Or that one video I saw a few years ago of a BW bot with perfect muta control getting stomped by a human player just massing goliaths and forcing it to take bad fights. The AlphaStar agent that played the Phoenix game vs MaNa showed really, really smart control, looking past the oppresively perfect precision and speed with which it controlled its units.
niteReloaded
Profile Blog Joined February 2007
Croatia5281 Posts
Last Edited: 2019-01-27 20:31:39
January 27 2019 20:28 GMT
#567
On January 27 2019 23:14 Haukinger wrote:
It's funny how people deem a korean terran exploiting marines with superior mechanical skill the highest level of play but if an AI uses blink stalkers it's just stupid with high apm.

It's not funny.
Super, human micro requires very rare talent and huge amounts of practice.
Super, AI micro requires coding it without APM limits.


Strategy, metagame, balance are all built on the capacity and potential of individual units. Unit's utility is based on what a player can do with it.
What a human player can do with a unit is vastly different from what a AI can do with it.

Based on that, AI uses different tactics, different strategies, almost a different game.

A human can't click 1500 in a minute, AlphaStar did.

StarCraft is a realtime strategy game where speed is very important. They made a bot which has, at crucial times, played at least 2X faster than any human can.

It's like building a sports playing bot which can run 2x faster. Suddenly, all tactics and strategies are affected and the game is almost not the same.

---
DeepMind's only mistake is that they didn't put a proper peak-APM limit.

Frankly, I would love to see games where AlphaStar is limited to 50 peak APM, then more games where the limit is 100 peak apm, then 150 etc.
Would love to see the strategies and tactics used.
Charoisaur
Profile Joined August 2014
Germany15925 Posts
January 27 2019 21:06 GMT
#568
On January 27 2019 23:14 Haukinger wrote:
It's funny how people deem a korean terran exploiting marines with superior mechanical skill the highest level of play but if an AI uses blink stalkers it's just stupid with high apm.

it's not stupid, it's just worthless if we want to know how good the AI is strategically if it has unlimited apm because then it can win with anything. If the goal would be to just beat a human no matter how there would be no need for the Deepmind team to take this on, a regular micro bot would be enough.
Many of the coolest moments in sc2 happen due to worker harassment
Xamo
Profile Joined April 2012
Spain877 Posts
January 28 2019 00:09 GMT
#569
I have not seen this posted elsewhere... this is MaNa's personal experience about DeepMind Starcraft 2 demonstration:

My life for Aiur. You got a piece of me, baby. IIIIIIiiiiiii.
Polypoetes
Profile Joined January 2019
20 Posts
Last Edited: 2019-01-28 01:14:29
January 28 2019 01:07 GMT
#570
It is like Jasper_Ty says. The more APM available to your bot, the harder it is for it to work correctly. Who knows what tricks the AI uses, or could use, vs other 'absurd no mistake' APM agents and that it wasn't using vs Mana or TLO?

I understand it is not interesting for you personally to improve your game. But we have had this discussion in the community ever since AlphaGo how long it would take for an AI to play properly and to beat top humans. I remember discussing with programmers who talked about how difficult of a problem it is to simulate combat before they make a move. Because that is what they tried to put into the BWAPI AIs. You see a game state, you see your units, you see the opponent units, you predict how the units will move and what they will attack, you calculate how much each side will lose, and then you know if this engagement is favorable.

So when people say that they consider the AI microing like crazy 'trivial', I just have to laugh.

Personally, I would like to see an unlimited APM AI. And I would love to see an AI play SC BW, because it seems to me that SC2 is way more straightforward and SC BW is way more on a knife edge and subtle. Maybe it is me being a former SC BW player, but I have this feeling that any play by an AI is way more exploitable in SC BW than in SC2.

Honestly, this reminds me of the debate we had back when SC2 was announced and when we had this influx of AoE, Civ, and C&C players arguing that slowing down the game, or making the interface easier, would give the game a richer game of strategy. The game is not what you think it is. Imagine an AI figuring out how to play Civ3. It is just calculating which builds to build first under which circumstances. It is boring. There is one solution. You calculate what it is, and you just carry it out.

Let's not forget that unlike Go or Chess, RTS games are convergent towards the end.
Kafka777
Profile Joined December 2015
361 Posts
January 28 2019 02:06 GMT
#571
To sum it up. They created AI that basicly can beat everyone with a drone rush. But strategically it has less brains than a monkey. Message to news - we beat best Sc2 players.
fededevi
Profile Joined April 2018
Italy45 Posts
January 28 2019 11:20 GMT
#572
Alphastar is impressive, it really is. But the strategies it came up with are not that interesting because humans cannot replicate them. The best plan is just the best plan you can execute correctly. And in order to make the AI come up with strategies that are useful to humans you will need to have a good human model to "limit" the AI in the correct way.

A simple APM limit is not gonna do the trick, sometimes a player can have huge "useful" apm spikes so you will need a model that account for that and many other things.

If you limit the AI too much, the AI will have to find workarounds to it's sub optimal micro by finding strategies that are sub-optimal for a player.
If you do not limit the AI enough, the AI will come up with strategies that players will never be able to execute.
Not to mention that each human player is different.

Still it would be very interesting to see if the AI can come up with good/decent "very-low-micro" strategies.

mierin
Profile Joined August 2010
United States4943 Posts
January 28 2019 13:32 GMT
#573
Mana literally "replicated" the AI strategy of over-probing on one base...
JD, Stork, Calm, Hyuk Fighting!
fededevi
Profile Joined April 2018
Italy45 Posts
January 28 2019 14:28 GMT
#574
That one was interesting indeed. In fact the macro part of the game is not so "apm-limited".
But we do not know if that strategy is actually good for human players too because the AI is playing with different limitations. If over-probing was done by all agents on different APM limits then it would be an indication that it is a good idea to always build more probes. With "strategy" I mean the whole game strategy, you can't take out a single thing (like over-probing) and discard the rest, because everything is connected in a single game.

For example the agent could be so good at winning with stalkers micro, that the only way to beat it is somehow to kill a lot of its probes. In this condition the agent would improve its chance of winning having a less than ideal number of probes. It would still have enough stalkers to win the engagements thanks to its god-like micro, but he would not lose to other kind of attacks to the mineral line. But if he was not so good at microing then it would have better chances by expanding its economy faster.

This is just an example, I don't know if this is the case.
Grumbels
Profile Blog Joined May 2009
Netherlands7031 Posts
January 28 2019 20:38 GMT
#575
An agent isn’t confident in its stalker micro, as it’s training vs bots, not humans. It won’t think it can win a four vs five stalker fight vs a human player because it is used to losing to other agents in such situations.

This is a common “weakness” to AI’s: a complete inability to take the psychology of the opponent into consideration. Agents tend to converge to the same cautious, conservative style which shows maximum respect to the opponent.

E.g. if you had to play a match versus an AI, it might be the case that if you just made blink stalkers it would vastly overestimate the strength of your army, because it doesn’t know you can’t actually micro blink stalkers at 1500 APM. And as a result it wouldn’t take any engagements.

This is why chess engines have a “contempt” setting programmed into them, which forces them to make sub-optimal moves which nevertheless increase winning chances against weaker opponents. This is a must in any sort of tournament or league play in chess, since it avoids draws.

I don’t know how you would program “contempt” into AlphaStar, other than training it versus agents with handicapped APM, such that it would develop more confidence in its micro, in order to better approach match conditions vs humans. But of course this doesn’t have any scientific value, only competitive utility.
Well, now I tell you, I never seen good come o' goodness yet. Him as strikes first is my fancy; dead men don't bite; them's my views--amen, so be it.
niteReloaded
Profile Blog Joined February 2007
Croatia5281 Posts
January 29 2019 20:25 GMT
#576
On January 29 2019 05:38 Grumbels wrote:
An agent isn’t confident in its stalker micro, as it’s training vs bots, not humans. It won’t think it can win a four vs five stalker fight vs a human player because it is used to losing to other agents in such situations.

This is a common “weakness” to AI’s: a complete inability to take the psychology of the opponent into consideration. Agents tend to converge to the same cautious, conservative style which shows maximum respect to the opponent.

E.g. if you had to play a match versus an AI, it might be the case that if you just made blink stalkers it would vastly overestimate the strength of your army, because it doesn’t know you can’t actually micro blink stalkers at 1500 APM. And as a result it wouldn’t take any engagements.

This is why chess engines have a “contempt” setting programmed into them, which forces them to make sub-optimal moves which nevertheless increase winning chances against weaker opponents. This is a must in any sort of tournament or league play in chess, since it avoids draws.

I don’t know how you would program “contempt” into AlphaStar, other than training it versus agents with handicapped APM, such that it would develop more confidence in its micro, in order to better approach match conditions vs humans. But of course this doesn’t have any scientific value, only competitive utility.

You release it on the ladder, with 1000s of instances playing vs humans 24/7.
Machine learning and genetic algorithms (or whatnot) would cause bots to start taking some smaller sub optimal moves, just to gauge the reactions of the opponent.
As the bots have played just other bots with same 'mechanical' capabilities, the ability to gauge opponent's mechanical skill wasn't necessary.
Cyro
Profile Blog Joined June 2011
United Kingdom20285 Posts
Last Edited: 2019-01-30 00:12:37
January 30 2019 00:11 GMT
#577
You release it on the ladder, with 1000s of instances playing vs humans 24/7.


There are not even that many humans on the ladder
"oh my god my overclock... I got a single WHEA error on the 23rd hour, 9 minutes" -Belial88
Deimos
Profile Joined June 2009
Mexico134 Posts
January 30 2019 02:39 GMT
#578
For me the most relevant thing its that the AI can't did anything new. It only got for metabuilds and won because its leve of ejecution and decision.
Acrofales
Profile Joined August 2010
Spain17976 Posts
January 30 2019 07:08 GMT
#579
On January 30 2019 09:11 Cyro wrote:
Show nested quote +
You release it on the ladder, with 1000s of instances playing vs humans 24/7.


There are not even that many humans on the ladder

Especially not at GM level. Pretty sure it's already good enough that even knowing its weaknesses it'll still beat diamond players on mechanics alone.
Plopus
Profile Joined November 2014
Switzerland112 Posts
Last Edited: 2019-01-30 09:09:07
January 30 2019 09:05 GMT
#580
Prev 1 27 28 29 30 Next All
Please log in or register to reply.
Live Events Refresh
Next event in 5h 34m
[ Submit Event ]
Live Streams
Refresh
StarCraft 2
Nina 220
Ketroc 47
StarCraft: Brood War
Leta 256
Snow 25
Noble 24
ajuk12(nOOB) 20
Icarus 11
Dota 2
monkeys_forever937
League of Legends
JimRising 738
Counter-Strike
Stewie2K689
Super Smash Bros
Mew2King66
Other Games
summit1g13607
shahzam1159
hungrybox997
WinterStarcraft430
C9.Mang0316
ViBE228
NeuroSwarm66
Trikslyr55
NotJumperer4
Organizations
Other Games
gamesdonequick2344
StarCraft 2
Blizzard YouTube
StarCraft: Brood War
BSLTrovo
sctven
[ Show 17 non-featured ]
StarCraft 2
• practicex 41
• intothetv
• IndyKCrew
• sooper7s
• AfreecaTV YouTube
• Migwel
• LaughNgamezSOOP
• Kozan
StarCraft: Brood War
• Azhi_Dahaki15
• STPLYoutube
• ZZZeroYoutube
• BSLYoutube
League of Legends
• Doublelift4383
• Rush1615
• Lourlo1160
• Stunt354
Other Games
• Scarra5227
Upcoming Events
The PondCast
5h 34m
OSC
8h 34m
WardiTV European League
11h 34m
Fjant vs Babymarine
Mixu vs HiGhDrA
Gerald vs ArT
goblin vs MaNa
Jumy vs YoungYakov
Replay Cast
19h 34m
Epic.LAN
1d 7h
CranKy Ducklings
2 days
Epic.LAN
2 days
CSO Contender
2 days
BSL20 Non-Korean Champi…
2 days
Bonyth vs Sziky
Dewalt vs Hawk
Hawk vs QiaoGege
Sziky vs Dewalt
Mihu vs Bonyth
Zhanhun vs QiaoGege
QiaoGege vs Fengzi
Sparkling Tuna Cup
3 days
[ Show More ]
Online Event
3 days
BSL20 Non-Korean Champi…
3 days
Bonyth vs Zhanhun
Dewalt vs Mihu
Hawk vs Sziky
Sziky vs QiaoGege
Mihu vs Hawk
Zhanhun vs Dewalt
Fengzi vs Bonyth
Esports World Cup
5 days
ByuN vs Astrea
Lambo vs HeRoMaRinE
Clem vs TBD
Solar vs Zoun
SHIN vs Reynor
Maru vs TriGGeR
herO vs Lancer
Cure vs ShoWTimE
Esports World Cup
6 days
Liquipedia Results

Completed

JPL Season 2
RSL Revival: Season 1
Murky Cup #2

Ongoing

BSL 2v2 Season 3
Copa Latinoamericana 4
Jiahua Invitational
BSL20 Non-Korean Championship
Championship of Russia 2025
FISSURE Playground #1
BLAST.tv Austin Major 2025
ESL Impact League Season 7
IEM Dallas 2025
PGL Astana 2025
Asian Champions League '25
BLAST Rivals Spring 2025
MESA Nomadic Masters

Upcoming

CSL Xiamen Invitational
CSL Xiamen Invitational: ShowMatche
2025 ACS Season 2
CSLPRO Last Chance 2025
CSLPRO Chat StarLAN 3
BSL Season 21
K-Championship
RSL Revival: Season 2
SEL Season 2 Championship
uThermal 2v2 Main Event
FEL Cracov 2025
Esports World Cup 2025
Underdog Cup #2
ESL Pro League S22
StarSeries Fall 2025
FISSURE Playground #2
BLAST Open Fall 2025
BLAST Open Fall Qual
Esports World Cup 2025
BLAST Bounty Fall 2025
BLAST Bounty Fall Qual
IEM Cologne 2025
TLPD

1. ByuN
2. TY
3. Dark
4. Solar
5. Stats
6. Nerchio
7. sOs
8. soO
9. INnoVation
10. Elazer
1. Rain
2. Flash
3. EffOrt
4. Last
5. Bisu
6. Soulkey
7. Mini
8. Sharp
Sidebar Settings...

Advertising | Privacy Policy | Terms Of Use | Contact Us

Original banner artwork: Jim Warren
The contents of this webpage are copyright © 2025 TLnet. All Rights Reserved.