History Main / ArtificialBrilliance

15th Aug '17 12:11:40 PM lalalei2001
Is there an issue? Send a Message


[[folder:Card Game]]

to:

[[folder:Card Game]]Games]]



** In ''Videogame/YuGiOhReshefOfDestruction'', the AI will use stalling tactics, multiplying monsters, and clever spells and trap cards.
* For Wizard of the Coast's ''MagicTheGathering'' video game Duel of the Planeswalkers the programmers had a very difficult task in programming an AI that could deal with all the rule-changing cards of the game itself (and there are possibly more of them than in any other card game because of Magic's "Golden Rule", which states that the text of a card takes priority over the rest of the rules). Seemingly they succeeded.

to:

** In ''Videogame/YuGiOhReshefOfDestruction'', ''VideoGame/YuGiOhReshefOfDestruction'', the AI will use stalling tactics, multiplying monsters, and clever spells and trap cards.
* For Wizard of the Coast's ''MagicTheGathering'' ''VideoGame/MagicTheGathering'' video game Duel of the Planeswalkers the programmers had a very difficult task in programming an AI that could deal with all the rule-changing cards of the game itself (and there are possibly more of them than in any other card game because of Magic's "Golden Rule", which states that the text of a card takes priority over the rest of the rules). Seemingly they succeeded.
22nd Jul '17 9:29:44 PM TotemicHero
Is there an issue? Send a Message


** See especially [[FanWorks/GalacticCivilizationsTFrancisVolOne this]] AfterActionReport, which goes so far as to cause Artificial ''FridgeBrilliance''. In the mid-to-end-game, the player was up against an alliance of a warmonger, a diplomat and an average race. The warmonger could have crushed him easily, but didn't. In fact, he even used his massive empire's voting power to punch through a galactic motion for peace against all three other races. The reason for this uncharacteristic behaviour: If the player had been defeated, it would have instantly resulted in an Alliance Victory, meaning a victory primary for the diplomat race that engineered the alliance. The goal of the warmonger faction was to keep the player alive until it was strong enough to break from the alliance and conquer both of its former partners at once - and the diplomat even anticipated this betrayal and built up his forces explicitly to be strong against his own ally. At which point the player happened to play "[[DarkHorseVictory lucky third party]]" by achieving a [[AscendToAHigherPlaneOfExistence Technological Victory]]. (The developers of the game have shed doubt on this interpretation, saying that the AI was not programmed to be ''that'' intelligent.)
20th Jul '17 8:29:22 PM Tacitus
Is there an issue? Send a Message


** See especially [[http://www.computerandvideogames.com/article.php?id=161570&site=pcg this]] AfterActionReport, which goes so far as to cause Artificial ''FridgeBrilliance''. In the mid-to-end-game, the player was up against an alliance of a warmonger, a diplomat and an average race. The warmonger could have crushed him easily, but didn't. In fact, he even used his massive empire's voting power to punch through a galactic motion for peace against all three other races. The reason for this uncharacteristic behaviour: If the player had been defeated, it would have instantly resulted in an Alliance Victory, meaning a victory primary for the diplomat race that engineered the alliance. The goal of the warmonger faction was to keep the player alive until it was strong enough to break from the alliance and conquer both of its former partners at once - and the diplomat even anticipated this betrayal and built up his forces explicitly to be strong against his own ally. At which point the player happened to play "[[DarkHorseVictory lucky third party]]" by achieving a [[AscendToAHigherPlaneOfExistence Technological Victory]]. (The developers of the game have shed doubt on this interpretation, saying that the AI was not programmed to be ''that'' intelligent.)

to:

** See especially [[http://www.computerandvideogames.com/article.php?id=161570&site=pcg [[FanWorks/GalacticCivilizationsTFrancisVolOne this]] AfterActionReport, which goes so far as to cause Artificial ''FridgeBrilliance''. In the mid-to-end-game, the player was up against an alliance of a warmonger, a diplomat and an average race. The warmonger could have crushed him easily, but didn't. In fact, he even used his massive empire's voting power to punch through a galactic motion for peace against all three other races. The reason for this uncharacteristic behaviour: If the player had been defeated, it would have instantly resulted in an Alliance Victory, meaning a victory primary for the diplomat race that engineered the alliance. The goal of the warmonger faction was to keep the player alive until it was strong enough to break from the alliance and conquer both of its former partners at once - and the diplomat even anticipated this betrayal and built up his forces explicitly to be strong against his own ally. At which point the player happened to play "[[DarkHorseVictory lucky third party]]" by achieving a [[AscendToAHigherPlaneOfExistence Technological Victory]]. (The developers of the game have shed doubt on this interpretation, saying that the AI was not programmed to be ''that'' intelligent.)



** Human enemies will make good use of cover, and if you do the same will try to flush you out with grenades. If one of their better-equipped comrades gets killed, they'll loot their weapon and use it themselves. And when a group of Raiders or Gunners attacks one of your settlements, they'll exploit any holes in your defenses and try to disable your generators first, to shut down your sentry guns.
** It's also a lot harder to use stealth in this game. If you snipe at people from a concealed position, they'll take blind potshots back in an attempt to spook you into fleeing while their friends move in to sweep the area, or just lob a missile or Mini-Nuke into where they think you're shooting from. If you're creeping around and backstabbing your targets, they'll search every dark corner for you. In any case, if anyone survives an encounter with you or comes across your handiwork, they'll run off and tell their friends to be on the alert.

to:

** Human enemies will make good use of cover, and if you do the same same, will try to flush you out with grenades. If one of their better-equipped comrades gets killed, they'll loot their weapon and use it themselves. If there's an unoccupied suit of power armor with a fusion core loaded, they'll happily jump in for the added protection. And when a group of Raiders or Gunners attacks one of your settlements, they'll exploit any holes in your defenses and try to disable your generators first, to shut down your sentry guns.
** It's also a lot harder to use stealth in this game. If you snipe at people from a concealed position, they'll take blind potshots back in an attempt to spook you into fleeing moving and revealing yourself, while their friends move in to sweep the area, or area - assuming they don't just lob a missile or Mini-Nuke into where they think you're shooting from. If you're creeping around and backstabbing your targets, they'll search every dark corner for you. In any case, if anyone survives an encounter with you or comes across your handiwork, they'll run off and tell their friends to be on the alert.
9th Jul '17 2:30:53 PM rjd1922
Is there an issue? Send a Message


Of course, it's a balancing act between an AI that is bad at the game and an AI that is too good at the game. The trick is allowing the AI to make human-like mistakes while also allowing it to have human-like brilliance. After all, in a first person shooter, the AI isn't really playing the game in the same way a human does. They don't actually have a mouse/keyboard to manipulate or have to watch a monitor. Thus it's an easy task to make an AI that always knows where you are and can hit you perfectly; it's not so easy to make an AI that can act like it doesn't know where you are and can act like it has reflexes. And as some of the examples below note (see the Half-Life example about greatest threat), sometimes being smart makes it dumb.

to:

Of course, it's a balancing act between an AI that is bad at the game and an AI that is too good at the game. The trick is allowing the AI to make human-like mistakes while also allowing it to have human-like brilliance. After all, in a first person shooter, the AI isn't really playing the game in the same way a human does. They don't actually have a mouse/keyboard to manipulate or have to watch a monitor. Thus it's an easy task to make an AI that always knows where you are and can hit you perfectly; it's not so easy to make an AI that can act like it doesn't know where you are and can act like it has reflexes. And as some of the examples below note (see the Half-Life ''Half-Life'' example about greatest threat), sometimes being smart makes it dumb.
9th Jul '17 11:56:51 AM Tacitus
Is there an issue? Send a Message

Added DiffLines:

* ''VideoGame/Fallout4'' features much improved enemy combat AI compared to earlier games in the series.
** Human enemies will make good use of cover, and if you do the same will try to flush you out with grenades. If one of their better-equipped comrades gets killed, they'll loot their weapon and use it themselves. And when a group of Raiders or Gunners attacks one of your settlements, they'll exploit any holes in your defenses and try to disable your generators first, to shut down your sentry guns.
** It's also a lot harder to use stealth in this game. If you snipe at people from a concealed position, they'll take blind potshots back in an attempt to spook you into fleeing while their friends move in to sweep the area, or just lob a missile or Mini-Nuke into where they think you're shooting from. If you're creeping around and backstabbing your targets, they'll search every dark corner for you. In any case, if anyone survives an encounter with you or comes across your handiwork, they'll run off and tell their friends to be on the alert.
** Low-level, generic Raiders suffer from ArtificialStupidity, and do things like hide behind [[EveryCarIsAPinto derelict cars]] or dismiss sounds as [[TheGuardsMustBeCrazy "probably nothing."]] Raider Veterans will not only ''not'' do that, but ''chew out their underlings for being so stupid.''
** Even the wasteland's mutated wildlife can be pretty smart. Mirelurks will cover their [[AttackItsWeakPoint vulernable faces]] with their claws as they scuttle towards you. Feral Mongrels will attack like packs of real dogs, circling around you and tearing at your flanks. Deathclaws will weave from side-to-side as they charge to foil your aim. And they're all smart enough to realize when you've climbed up or crawled into a place they can't reach, and will run off rather than let you shoot them with impunity, at least until they find another way to get at you.
19th Jun '17 1:08:00 AM ImpudentInfidel
Is there an issue? Send a Message


* It was believed that it would take a long time for computers to become very good at {{Go}} (the game has a far greater number of possible layouts than TabletopGame/{{Chess}}, so computers can't just use the "perfect lookup" method), but in March 10, 2016, Google's [=AlphaGo=] beat Lee Sedol, a world champion in Go. It is noted, however, that Mr. Sedol had been beaten before this by another chinese player, so he was only a ''former'' champion.

to:

* It was believed that it would take a long time for computers to become very good at {{Go}} (the game has a far greater number of possible layouts than TabletopGame/{{Chess}}, so computers can't just use the "perfect lookup" method), but in March 10, 2016, Google's [=AlphaGo=] beat Lee Sedol, a world champion in Go. It is noted, however, that Mr. Sedol had been beaten before this by another chinese player, so he was only a ''former'' champion. It beat the current champion in May of 2017.
17th Jun '17 1:49:16 AM syckls
Is there an issue? Send a Message


* Computers have gotten really, [[UpToEleven really]] good at chess. Essentially, it is impossible to beat a computer at chess if it is using all of its ability. The reason being that computers are now advanced enough to store ''every possible move in chess'' in their memories, meaning all they have to do is look up the perfect counter for whatever move one uses against them.

to:

* Computers have gotten really, [[UpToEleven really]] good at chess. Essentially, it is impossible to beat a computer at chess if it is using all of its ability. The reason being that computers are now advanced enough to store ''every possible move in chess'' in their memories, meaning all they have to do is look up the perfect counter for whatever move one uses against them.
10th Jun '17 7:48:02 AM SJMistery
Is there an issue? Send a Message

Added DiffLines:

** The Battle Tree opponents have FINNALLY learnt to switch their pokemon around when they have no attacks that will deal at least neutral damage. They still try to spam boosting moves before realizing it won't work, and they still fail to catch on against your own strategies, like when you use two-turn attacks like Fly and Bounce and they have Protect, but it's a good start. They even make use mega pokemon and Z-moves right away when they need them!
2nd Jun '17 7:33:31 AM Morgenthaler
Is there an issue? Send a Message


* ''AIWarFleetCommand'', a seemingly innocuous indie 2D space RTS made by a single person. But that doesn't stop the AI from actually understanding flanking tactics, creating distractions for the human player, utilizing hit and run warfare...

to:

* ''AIWarFleetCommand'', ''VideoGame/AIWarFleetCommand'', a seemingly innocuous indie 2D space RTS made by a single person. But that doesn't stop the AI from actually understanding flanking tactics, creating distractions for the human player, utilizing hit and run warfare...
14th May '17 4:52:05 PM nombretomado
Is there an issue? Send a Message


* In 1981, and then again in 1982, Douglas Lenat [[http://www.newyorker.com/reporting/2009/05/11/090511fa_fact_gladwell?currentPage=all tested his learning program]], [[http://en.wikipedia.org/wiki/Eurisko Eurisko]], in a {{Traveller}}: Trillion Credit Squadron tournament. Eurisko [[LevelGrinding simulated thousands of battles]], [[LoopholeAbuse found unconventional ship configurations and methods]], and defeated all comers. Twice. In a row. Even with notable rule changes. Eurisko could have done it a third time, but Lenat decided to retire it from the tournament, since if the program had won a 3rd time, it would be the last such tournament.

to:

* In 1981, and then again in 1982, Douglas Lenat [[http://www.newyorker.com/reporting/2009/05/11/090511fa_fact_gladwell?currentPage=all tested his learning program]], [[http://en.wikipedia.org/wiki/Eurisko Eurisko]], in a {{Traveller}}: ''TabletopGame/{{Traveller}}'': Trillion Credit Squadron tournament. Eurisko [[LevelGrinding simulated thousands of battles]], [[LoopholeAbuse found unconventional ship configurations and methods]], and defeated all comers. Twice. In a row. Even with notable rule changes. Eurisko could have done it a third time, but Lenat decided to retire it from the tournament, since if the program had won a 3rd time, it would be the last such tournament.
This list shows the last 10 events of 468. Show all.
http://tvtropes.org/pmwiki/article_history.php?article=Main.ArtificialBrilliance