The document discusses the concept of autonomy in artificial intelligence (AI) systems, highlighting their ability to make independent decisions, learn from experience, and adapt to new information. It also addresses concerns related to accountability, privacy, and biases in AI, as well as the implications of autonomous systems in various fields, including military applications. Additionally, the document touches on the challenges of ensuring transparency and reliability in AI decision-making processes.
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0 ratings0% found this document useful (0 votes)
16 views18 pages
EAI Notes
The document discusses the concept of autonomy in artificial intelligence (AI) systems, highlighting their ability to make independent decisions, learn from experience, and adapt to new information. It also addresses concerns related to accountability, privacy, and biases in AI, as well as the implications of autonomous systems in various fields, including military applications. Additionally, the document touches on the challenges of ensuring transparency and reliability in AI decision-making processes.
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 18
Detrne fy Exelatrs Fa\louning
§] Automonay 1h abil:
{J Autonomy tes ATE
Auionemy relerc to the obiiine of cam
artificial inrelligence. cystem tol
make decisions inclepenctendiy Aso We Co micros
human interverton,
—Autonomous AT systems can analyze frformation.
= learn fron, oak experience .
= Different AT sustern ha. thy,
Quronamy s
= Autohomous Su ctenne tuck selE dretvitg “care
acidlynes,- Can “Prake Aecisiong tn AieRcult Sthvety
ii) Adeptabilthy ip AT i=
Adap' atl ity refers tp the systems Capacity
i a apis t 2 3
SPD adaptable a sycern can leann Gran Seperience,
yefine tke models ancl enhance decislorn paling
aki\iky.
=
—. Askerackiyas refers +o Commauntcottoni
2 Collahoxnitnn bohneen Uumans
fox letter Uedensanad Ling
Why Beep Learni ed a Black Bok 2
Explain witty one example. ___
Peep Learning tx oflem reCFered
Peis bemuse | of the cy t
& urdexsbareing aus model arives at ptr
Aecisiome . fick
= The Init hol oxiding 10 Nahe
da ma ceNaly, “y@eceumanten Pe
me wy
Tn cational prgtamntng reaching Learnt ps
maodels the ‘elation helvecn ir ae |
is Avrarsparent 4 sdetstascobsle:. = a
= pe fou
abd trainable PArOMmeie AS A pesubh
Jecision caadetng paces _becames Mighty comeley,*) Privacy Concerns
SAO =
= AT sysienn Gather and analyze Vou arroumt
of data.
= _Frobectior ch a tinclivi vi
becomes a signiFicank choad Ange
iit] Aransparences § Account bi\thy
== Complex mode of AL Systems taker ie
AifSiculy to. onde novela Vice eae
= Ft should be accourmkab in FE of ervorx
and _ bio«
W) @eurtky Risks iney Bee Seity Ccomc Aecicior
= Ek THe she, is situadion rhak involve
harta tn __pasiengers ib Cure ox people 1
cubs cave cake ing oh rend. oo ae
a this raises Moral quectiorm
i) Liat Shy Dcourtabs thy
é eke nining recponsibtlity in te eee
Ack Or shen nt re. ae
= ho should be hetd accountable
mmabufocture , We sofhog
bevel op ee ee: Mev =
athe “ngsuiiyebons eas And 4
Piearckivcnatoiwsallem ode howe \ekely kn be be a
i Pieoal aboot i ae —
ay teflecks O. geheloe kee eo he ae
7
c ies alone and femats, neglactrs EE
- | __ she Fack bck pecple of dy gender ne
= || Seek Solon + servicer eet 4
= This dype of bi r position bt ag
Se removing stich, |
> | biases =
whic OK C
Pen OE. ALPS take, beast ake §
Neer pastor ne
vie | Fironcigl Tramacks a=
Sirarsavhdh an GPan ghonpe apr oho
et al Es
a iL wit) Acuthentsccs'on hae Acuiteo 2S ganas
gener coos, of date elated) to
z \ogims, biromesrics & ecarciy.
5 a)SL Veack r Ke wed ourkcomes
\\ Sampling Bias t
BIE Scr iien UShens ‘dail nitceue tire
“
=_One_ginirp oth, rote Gnd cre cath less:
|
ir.
WW) Oherver Biact
= Iniveduced whem person ccllecking dodo
ev aes 3 tae
(nGuence pes
WW) Measurmony Big =
= trove due +o ihstrumentic or meacuements.
Y) MgortHaric Bias $=
= OClUrseae he cleplo
En ire, isc
Pverse
— Reg Ay AU cA
Aateuek: “Ga bta«
Bee deine eollance Woe Arai mg
BeOrcnme bias perectfon tole
BRS and Glow guideline
fx datu collection
EE be Sct opel y reset ghd ig
J
Lechmiquer.
+ {Ad oeReal teftme eperdising
A Geal-Aih-e _operadsing Syclern fo Gr _Opemie
Aan Fo
dead \ine
& Chacackericic Z ~
4) Peherm@esuic Behavior =
= G@ucanke tod Cerkain toate wilt be Completsl!
Win We ehownm Aime .
‘J Tost sched ree -
=)
= Tost cled Ming is_ 9 axe
higher axtaity tact tn execute Srrek
1K) Reat- ‘Aree choc =
= Th includes eal ime Clack oy Aftee ap
deag Aire centsve, _operatin ine
Sw pe a ee
NJ) Relrabiithy t= |
aaa aA vagely reliable,= used tn Medical equtpmente
= Potent fr pitoxdng Syctern
= Cancer / Tumer detectionAdvantages t= eee
Suiom fa CU oy
= Aitrrobeia Glens edie ee
paake dedisions Gister stam bhunsape
= Deploying autonomous Capone com redude
Peers hursnn stl detec 0. leteeeer eee
= nee —
24 [# wine fakian
= Pructo ino pn cudepn carn be scalable
@ > es
* Dis advantages —
Bou .ur jeanses pele oe
4 roorgl 2
= Whors shod be accountable Im cowe of
aoe Jararmage done Lays Auta pormoter Sy stern.
ie cmd decile electtocemea eam cL. |
by AA oO OE SSE S « i
Saeailencesce il