CSE1GDT - Lecture 9b - Artificial Stupidityx

muscleblouseAI and Robotics

Oct 19, 2013 (3 years and 11 months ago)

88 views

Artificial Stupidity

Paul Taylor 2010

Artificial Intelligence


What is it?

Artificial


Fake!

Intelligence

Latin:
intelligere:
-

‘to understand’



There are two general meanings derived from
this, one by the APA, and a second by a group
of intelligence researchers

APA

“Individuals differ from one another in their
ability to understand complex ideas, to adapt
effectively to the environment, to learn from
experience, to engage in various forms of
reasoning, to overcome obstacles by taking
thought.”


Source: Intelligence:
Knowns

and Unknowns,
1996, AP, February

Mainstream Science on Intelligence

“A very general mental capability that, among other
things, involves the ability to reason, plan, solve
problems, think abstractly, comprehend complex
ideas, learn quickly and learn from experience.

It is not merely book learning, a narrow academic
skill, or test
-
taking smarts.

Rather, it reflects a broader and deeper capability
for comprehending our surroundings

"catching
on", "making sense" of things, or "figuring out"
what to do”

Source: Wall Street Journal, December, 1994

What is the essence of these
statements?

APA: “Individuals differ from one another in their ability to
understand
complex ideas
, to
adapt effectively
to the environment, to
learn

from experience
, to engage in various forms of
reasoning
, to
overcome obstacles by
taking thought
.”

Int

Sci
: “A very general mental capability that, among other things,
involves the ability to
reason
,
plan
, solve problems,
think abstractly
,
comprehend complex ideas
,
learn

quickly and
learn from
experience.

It is not merely book learning, a narrow academic skill, or test
-
taking
smarts.

Rather, it reflects a broader and deeper
capability for comprehending
our surroundings

"catching on", "making sense" of things, or
"figuring out" what to do”




The Roots of AI

The Difference Engine


1822, Charles Babbage


It was a Polynomial Calculator


See LBP version

http://www.meccano.us/difference_engines/rde_1/DEFront.jpg

1837 The Analytical Engine


Effectively would have been a mechanical
computer


Babbage worked on it till he died in 1871


Work was abandoned in 1878

1950 The Turing Test

http://huehueteotl.files.wordpress.com/2007/
04/turing_test.gif


http://languagelog.ldc.upenn.edu/myl/llog/DilbertTuring.jpg

1990 The Loebner Prize


http://www.jabberwacky.com/chat
-
george


Symbolic Languages


http://library.wustl.edu/units/spec/rarebooks/
semeiology/arnold6.gif

Neat and Scruffy AI


http://www.goodlookingdog.com/uploads/Scr
uffy.jpg

http://www.greghoganphotography.com/nagoh/wp
-
content/uploads/2008/07/scruffy.jpg

Expert Systems


http://oopweb.com/Prolog/Documents/XSIP/V
olume/intro1
-
1.gif

http://richardbowles.tripod.com/cpp/projects/
proj3a.gif

Genetic Algorithms


http://www.c
-
sharpcorner.com/UploadFile/mgold/GeneticAlgorithm12032005044205AM/Images/GeneticAlgo
rithm.gif

http://folk.uio.no/oivindb/bilder/chicken.gif

Neural Networks (Connectionism)


http://aemc.jpl.nasa.gov/images/neural_netw
ork_browse2.jpg

http://computingassociate.com/images/neural
-
network.jpg

The Problem: Snails


A Snail has only 3 neurons!!!

http://idle.slashdot.org/story/10/06/01/13242
20/Snails
-
On
-
Methamphetamine

http://i199.photobucket.com/albums/aa199/d
k49/neural
-
network.jpg

The CYC Project cyc.com


This is a project designed to capture the commonsense of
humanity, ambitious: YES


It currently contains x00,000’s of terms and millions of
assertions.


The founder, Doug Lenat predicts that the project will need
around 100 million assertions before it can begin ‘reading’
literature.

http://64.202.120.86/upload/im
age/personal
-
column/israel
-
biniamini/one
-
more
-
meaning
-
of
-
computer
-
science/cyc.jpg

And that’s where we are at!


Google Images ‘AI’


See ‘Don’t Burn the
Cat’


http://thefutureofthings.com/column/1001/dont
-
burn
-
the
-
cat.html

http://aftermathnews.files.wordpress.com/200
9/02/forbidden_planet_robot1.jpg?w=287&h=
421

http://www.cs.umb.edu/~marc/cs271/AI.jpg

http://images.eonline.com/eol_images/Entire_
Site/20080523/300.terminator.052308.jpg

Video Game AI

Steve Rabin (Game AI) provides 2 definitions:


“Game AI is any technique that contributes to the
perceived intelligence of an entity, regardless of what’s
under the hood.”



“Techniques that generate artefacts that would normally
be created by the game developers (like
environment/layout, story, music, or
models/animations/textures).”


Video Game AI

When was the first AI introduced to gaming?

Video Game AI

When was the first AI
introduced to
gaming?


It all depends on your
classification of AI


Space Invaders
responded to the
player input...

http://www.reelmovienews.com/files/space
-
invaders
-
still.jpg

Halo AI


Chris Butcher


The character uses its AI "senses" to perceive the world
--

to detect what's going on around it.


The AI takes the raw information that it gets based on its
perception and interprets the data.


The AI turns that interpreted data into more processed
information


The AI makes decisions about what its actions should be
based on that information.


Then the AI figures out how it can best perform those
actions to achieve the desired result based on the physical
state of the world around it.


Halo 1 Decision Making

Only 4 states:


Idle


Guard/Patrol


Attack/Defend


Retreat


Tough Means Smart!


When you improve the AI relative to the
player, obviously the game gets tougher.


However, when you make the AI tougher (e.g.
with more health points), the players believe
they are smarter.


Balance Difficulty Levels

“If the game goes from easy to hard and back
again, players attribute the easy parts to poor
AI and the hard parts to evil level designers.”

Make Behaviours Even More Obvious

“Initially nobody noticed so we had to keep adding
clues to make it more obvious. By the time we
shipped we had made it so not only does
every
single

Grunt run away
every single

time an Elite is
killed.”


Even with the exaggerated animations and sounds,
they estimate that less than a third of the players
understood the connection.

Sensible Design Goals


Intelligible


The AI should be easy to
understand.


Interactive


All in
-
game actors should focus
on the player.


Unpredictable


Gameplay

should not
become repetitive.


Playtest

as Much as You Can


A very typical statement of Bungie



http://img.qj.net/uploads/articles_module2/7
4305/news_hb3_qjpreviewth.jpg

http://meekmok.com/sassy/images/playtestin
g_poster.jpg

Cover Up Weaknesses with Design

A good design always takes into account problems with AI and
gameplay
, and addresses these issues.

In
Halo

for example, two specific problems were fixed by
negatively reinforcing certain playing tactics:


Rushing certain levels with the assault rifle isn’t fun
because the levels end too quickly. To fix this,
Bungie

made
the AI more accurate the longer the player is out of cover.


Also, the AI isn’t very good at dealing with close targets
moving quickly, so the design was extended to give them
powerful melee attacks.


Details May Appear as Bugs

“Early versions of Halo’s AI had a short pause
between seeing you and recognizing that you
were an enemy. This was to allow people to
get the jump on the AI or retreat if they were
getting in over their heads.
Playtest

feedback
showed that players assumed this Feature was
a Bug and they all said it made the AI look
dumb.”

Other ‘tweaks’ in Halo

Strategic Spaces to Emphasize the AI:


Battle Area


Linear corridors aren’t much fun to
have battles in. Instead there needs to be lots of
interconnected space for flanking and other
tactical manoeuvres.


Killing Zones


Since large spaces can be chaotic,
the designers must establish an open area
between two positions with good cover. This
makes the battle feel more directed.


Other ‘tweaks’ in Halo

Tactical Scripting to Improve Battles:


An aggressive territory that it should occupy
by attacking.


A defensible location for the group to fallback
to.


Other ‘tweaks’ in Halo

Annotated Firing Positions to Control Battles:


In
Halo

the designers can add annotations to
the level to specify firing positions. This
provides better time
-
dependent control over
the combat

Other ‘tweaks’ in Halo

Emphasize the Tactical Goals of Squads:


“Players are naively confident in your AI
programmer. They will assume your AI knows
they are supposed to be protecting the tractor
beam power switch until you prove that they
are just standing around.”

Other ‘tweaks’ in Halo

Make it Easy to Control Friendly AI:


“Unfortunately the Player is often forced to be
a manager instead of a hero the end result is
that the Player often ignores this option.”

The friendly AI actors are certainly autonomous,
but they can still get into trouble with
grenades, etc. A few
Halo 3

reviews mention
these bugs in some way… Co
-
op AI is certainly
not an easy problem to solve.

The Halo 2 Combat Cycle


http://www.gamasutra.com/gdc2005/features
/20050311/figure01.jpg

Halo 2 Behaviour Tree


http://www.gamasutra.com/gdc2005/features/20050311/figure02.jpg

The Halo 2 AI Hierarchy


http://www.gamasutra.com/gdc2005/features
/20050311/figure07.jpg

Designing your own AI


Know the scale


Choose the simplest usable form


Implement the AI independent of the level
design


Have fun finding your bugs


The End


http://aigamedev.com/open/reviews/halo
-
ai/


http://electronics.howstuffworks.com/halo2
-
ai2.htm


http://www.bungie.net/images/Inside/publica
tions/presentations/publicationsdes/design/g
dc02_jaime_griesemer.pdf


http://www.gamasutra.com/view/feature/225
0/gdc_2005_proceeding_handling_.php