{"id":39055,"date":"2019-10-31T22:27:36","date_gmt":"2019-10-31T19:27:36","guid":{"rendered":"https:\/\/prohoster.info\/blog\/my-ne-mozhem-doveryat-ii-sistemam-postroennym-na-odnom-lish-glubokom-obuchenii\/"},"modified":"2019-10-31T22:27:36","modified_gmt":"2019-10-31T19:27:36","slug":"my-ne-mozhem-doveryat-ii-sistemam-postroennym-na-odnom-lish-glubokom-obuchenii","status":"publish","type":"post","link":"https:\/\/prohoster.info\/sq\/blog\/news\/my-ne-mozhem-doveryat-ii-sistemam-postroennym-na-odnom-lish-glubokom-obuchenii","title":{"rendered":"We cannot trust AI systems built solely on deep learning.","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

\"We
\n
\nKy ky\u00e7 i k\u00ebtij teksti nuk \u00ebsht\u00eb nj\u00eb rezultat i k\u00ebrkimit shkencor, por nj\u00eb nga shum\u00eb mendime n\u00eb lidhje me zhvillimin ton\u00eb m\u00eb t\u00eb af\u00ebrt teknologjik. Gjithashtu, \u00ebsht\u00eb nj\u00eb ftes\u00eb p\u00ebr diskutim.<\/i><\/p>\n

Gary Marcus, profesor n\u00eb Universitetin e New York-ut, \u00ebsht\u00eb i sigurt se m\u00ebsimi i thell\u00eb luan nj\u00eb rol t\u00eb r\u00ebnd\u00ebsish\u00ebm n\u00eb zhvillimin e AI. Por ai gjithashtu beson se nj\u00eb pasion i tepruar p\u00ebr k\u00ebt\u00eb metodologji mund t\u00eb \u00e7oj\u00eb n\u00eb diskreditimin e saj.<\/p>\n

N\u00eb librin e tij Rebooting AI: Building artificial intelligence we can trust<\/i> Marcus, me formim si neurolog, i cili ndihmoi n\u00eb nd\u00ebrtimin e nj\u00eb karriere n\u00eb k\u00ebrkime t\u00eb avancuara n\u00eb fush\u00ebn e AI, merret me aspektet teknike dhe etike. Nga pik\u00ebpamja teknologjike, m\u00ebsimi i thell\u00eb mund t\u00eb imitoj\u00eb me sukses zgjidhjen e problemeve perceptuese q\u00eb b\u00ebjn\u00eb truri yn\u00eb: p\u00ebr shembull, njohja e imazheve ose fjal\u00ebve. Por p\u00ebr zgjidhjen e \u00e7\u00ebshtjeve t\u00eb tjera, si kuptimi i bisedave apo p\u00ebrcaktimi i lidhjeve shkak\u00ebsore, m\u00ebsimi i thell\u00eb nuk \u00ebsht\u00eb i p\u00ebrshtatsh\u00ebm. P\u00ebr t\u00eb krijuar makina intelegjente m\u00eb t\u00eb avancuara, t\u00eb cilat mund t\u00eb zgjidhin nj\u00eb gam\u00eb m\u00eb t\u00eb gjer\u00eb problemesh \u2014 shpesh quhen inteliigjenc\u00eb artificiale e p\u00ebrgjithshme \u2014 m\u00ebsimi i thell\u00eb duhet t\u00eb kombinohet me metoda t\u00eb tjera.
\n<\/a><\/noindex>
\nN\u00ebse nj\u00eb sistem AI n\u00eb t\u00eb v\u00ebrtet\u00eb nuk kupton detyrat e tij ose bot\u00ebn p\u00ebrreth, kjo mund t\u00eb \u00e7oj\u00eb n\u00eb pasoja t\u00eb rrezikshme. Edhe ndryshimet m\u00eb t\u00eb vogla t\u00eb papritura n\u00eb mjedisin e sistemit mund t\u00eb \u00e7ojn\u00eb n\u00eb sjellje t\u00eb gabuara. K\u00ebto shembuj kan\u00eb qen\u00eb t\u00eb shumt\u00eb: identifikues t\u00eb shprehjeve t\u00eb pap\u00ebrshtatshme, t\u00eb cilat jan\u00eb leht\u00eb p\u00ebr t\u2019u mashtruar; sisteme p\u00ebr k\u00ebrkim t\u00eb pun\u00ebs q\u00eb vazhdimisht diskriminojn\u00eb; automjete pa shofer q\u00eb kan\u00eb pasur aksidente dhe ndonj\u00ebher\u00eb vrasin drejtuesin ose kalimtar\u00ebt. Krijimi i inteligjenc\u00ebs artificiale t\u00eb p\u00ebrgjithshme nuk \u00ebsht\u00eb thjesht nj\u00eb problem k\u00ebrkimor interesant, por ka nj\u00eb s\u00ebr\u00eb aplikimesh tep\u00ebr praktike.<\/p>\n

N\u00eb librin e tij, Marcus dhe bashkautorja e tij Ernest Davis promovojn\u00eb nj\u00eb rrug\u00eb tjet\u00ebr. Ata besojn\u00eb se jemi ende larg krijimit t\u00eb AI t\u00eb p\u00ebrgjithsh\u00ebm, por jan\u00eb t\u00eb sigurt se n\u00eb nj\u00eb moment ose nj\u00eb tjet\u00ebr, do t\u00eb arrijm\u00eb ta krijojm\u00eb at\u00eb.<\/p>\n

P\u00ebr \u00e7far\u00eb na nevojitet AI e p\u00ebrgjithshme? Versionet e specializuara tashm\u00eb jan\u00eb krijuar dhe sjellin shum\u00eb p\u00ebrfitime.<\/b><\/p>\n

Ka p\u00ebshpejton, dhe do ket\u00eb akoma m\u00eb shum\u00eb p\u00ebrfitime. Por ka shum\u00eb detyra q\u00eb nj\u00eb AI i specializuar thjesht nuk \u00ebsht\u00eb n\u00eb gjendje t'i zgjidh\u00eb. P\u00ebr shembull, kuptimi i t\u00eb folurit t\u00eb zakonsh\u00ebm, apo ndihma e p\u00ebrgjithshme n\u00eb bot\u00ebn virtuale, ose nj\u00eb robot q\u00eb ndihmon n\u00eb pastrimin dhe gatimin. K\u00ebto detyra jan\u00eb jasht\u00eb mund\u00ebsive t\u00eb AI-s\u00eb s\u00eb specializuar. Nj\u00eb pyetje tjet\u00ebr interesante praktike: a mund t\u00eb krijohet nj\u00eb makin\u00eb e sigurt pa pilot me ndihm\u00ebn e nj\u00eb AI t\u00eb specializuar? Eksperienca tregon q\u00eb nj\u00eb AI i till\u00eb ende ka shum\u00eb probleme me sjelljen n\u00eb situata anomale, madje dhe gjat\u00eb drejtimit, gj\u00eb q\u00eb e komplikon ndjesh\u00ebm situat\u00ebn.<\/p>\n

Mendoj se t\u00eb gjith\u00eb ne do t\u00eb donim nj\u00eb AI q\u00eb mund t\u00eb na ndihmoj\u00eb t\u00eb b\u00ebjm\u00eb zbulime t\u00eb reja n\u00eb mjek\u00ebsi. Nuk \u00ebsht\u00eb e qart\u00eb n\u00ebse teknologjit\u00eb aktuale do t\u00eb ishin t\u00eb p\u00ebrshtatshme p\u00ebr k\u00ebt\u00eb, pasi biologjia \u00ebsht\u00eb nj\u00eb fush\u00eb e komplikuar. Duhet t\u00eb jemi t\u00eb gatsh\u00ebm t\u00eb lexojm\u00eb shum\u00eb libra. Shkenc\u00ebtar\u00ebt kuptojn\u00eb lidhjet shkak-pasoj\u00eb n\u00eb nd\u00ebrveprimin e rrjeteve dhe molekulave, mund t\u00eb zhvillojn\u00eb teori p\u00ebr planet\u00eb dhe k\u00ebshtu me radh\u00eb. Megjithat\u00eb, me AI t\u00eb specializuar ne nuk mund t\u00eb krijojm\u00eb makina q\u00eb jan\u00eb n\u00eb gjendje p\u00ebr k\u00ebto zbulime. Nd\u00ebrsa me AI t\u00eb p\u00ebrgjithsh\u00ebm do t\u00eb ishim n\u00eb gjendje t\u00eb b\u00ebnim nj\u00eb revolucion n\u00eb shkenc\u00eb, teknologji dhe mjek\u00ebsi. M\u00eb duket shum\u00eb e r\u00ebnd\u00ebsishme t\u00eb vazhdojm\u00eb pun\u00ebn mbi krijimin e AI-s\u00eb s\u00eb p\u00ebrgjithshme.<\/p>\n

Duket se me \"t\u00eb p\u00ebrgjithshem\" n\u00ebnkupton nj\u00eb AI t\u00eb fuqish\u00ebm?<\/b><\/p>\n

Duke th\u00ebn\u00eb \"t\u00eb p\u00ebrgjithsh\u00ebm\", po n\u00ebnkuptoj se AI do t\u00eb jet\u00eb n\u00eb gjendje t\u00eb mendoj\u00eb dhe t\u00eb zgjidh\u00eb detyra t\u00eb reja n\u00eb fluturim. Ndryshe nga, le t\u00eb themi, Go, ku problemi nuk ka ndryshuar p\u00ebr 2000 vjet.<\/p>\n

AI e p\u00ebrgjithshme duhet t\u00eb jet\u00eb n\u00eb gjendje t\u00eb marr\u00eb vendime si n\u00eb politik\u00eb ashtu edhe n\u00eb mjek\u00ebsi. Kjo \u00ebsht\u00eb nj\u00eb analogji me aft\u00ebsin\u00eb njer\u00ebzore; \u00e7do person i arsyesh\u00ebm mund t\u00eb b\u00ebj\u00eb shum\u00eb gj\u00ebra. Merrni student\u00eb t\u00eb paekspertizuar dhe pas disa dit\u00ebsh i b\u00ebni t\u00eb punojn\u00eb praktikisht n\u00eb \u00e7do gj\u00eb, duke filluar nga nj\u00eb detyr\u00eb ligjore dhe duke p\u00ebrfunduar me at\u00eb mjek\u00ebsore. Kjo \u00ebsht\u00eb p\u00ebr shkak se ata kan\u00eb nj\u00eb kuptim t\u00eb p\u00ebrgjithsh\u00ebm t\u00eb bot\u00ebs dhe din\u00eb t\u00eb lexojn\u00eb, dhe p\u00ebr k\u00ebt\u00eb arsye mund t\u00eb kontribuojn\u00eb n\u00eb nj\u00eb gam\u00eb shum\u00eb t\u00eb gjer\u00eb aktivitetesh.<\/p>\n

Lidhja midis nj\u00eb intelekti t\u00eb till\u00eb dhe atij t\u00eb fuqish\u00ebm \u00ebsht\u00eb se nj\u00eb intelekt jo i fuqish\u00ebm, p\u00ebr shum\u00eb shanse, nuk do t\u00eb jet\u00eb n\u00eb gjendje t\u00eb zgjidh\u00eb detyra t\u00eb p\u00ebrgjithshme. P\u00ebr t\u00eb krijuar di\u00e7ka mjaft t\u00eb besueshme, e cila mund t\u00eb funksionoj\u00eb n\u00eb nj\u00eb bot\u00eb q\u00eb ndryshon vazhdimisht, ndoshta ju nevojitet t\u00eb pakt\u00ebn t\u00eb afroni n\u00eb inteligjenc\u00ebn e p\u00ebrgjithshme.<\/p>\n

Por tani, ne jemi aspak af\u00ebr k\u00ebsaj. AlphaGo mund t\u00eb luaj\u00eb shk\u00eblqyesh\u00ebm n\u00eb nj\u00eb tabel\u00eb 19×19, por duhet t\u00eb ritrainohet p\u00ebr t\u00eb luajtur n\u00eb nj\u00eb tabel\u00eb katrore. Ose merrni nj\u00eb sistem mesatar t\u00eb m\u00ebsimit t\u00eb thell\u00eb: ai mund t\u00eb njoh\u00eb nj\u00eb elefant, n\u00ebse \u00ebsht\u00eb ndri\u00e7uar mir\u00eb dhe tekstura e l\u00ebkur\u00ebs \u00ebsht\u00eb e dukshme. Por n\u00ebse shihet vet\u00ebm silueti i elefantit, sistemi me siguri nuk do t\u00eb jet\u00eb n\u00eb gjendje ta njoh\u00eb.<\/p>\n

N\u00eb librin tuaj p\u00ebrmendni se m\u00ebsimi i thell\u00eb nuk \u00ebsht\u00eb n\u00eb gjendje t\u00eb arrij\u00eb kapacitetet e AI-s\u00eb s\u00eb p\u00ebrgjithshme, pasi nuk \u00ebsht\u00eb n\u00eb gjendje t\u00eb realizoj\u00eb nj\u00eb kuptim t\u00eb thell\u00eb.<\/b><\/p>\n

N\u00eb kognitivistik\u00eb flitet p\u00ebr formimin e modeleve t\u00eb ndryshme kognitive. Un\u00eb q\u00ebndroj n\u00eb dhom\u00ebn e hotelit dhe kuptoj q\u00eb atje \u00ebsht\u00eb nj\u00eb dollap, atje \u00ebsht\u00eb nj\u00eb krevat, atje \u00ebsht\u00eb nj\u00eb televizor q\u00eb \u00ebsht\u00eb n\u00eb m\u00ebnyr\u00eb t\u00eb pazakont\u00eb i varur. E di t\u00eb gjitha k\u00ebto objekte, nuk i identifikoj vet\u00ebm. Gjithashtu, kuptoj se si jan\u00eb t\u00eb lidhura nd\u00ebrmjet tyre. Kam ide p\u00ebr funksionimin e bot\u00ebs p\u00ebrreth. Ato nuk jan\u00eb perfekte. Mund t\u00eb jen\u00eb t\u00eb gabuara, por jan\u00eb t\u00eb mira. Dhe mbi k\u00ebt\u00eb baz\u00eb, b\u00ebj shum\u00eb konkluzione, t\u00eb cilat b\u00ebhen udh\u00ebzime p\u00ebr veprimet e mia t\u00eb p\u00ebrditshme.<\/p>\n

Ekstremi tjet\u00ebr \u00ebsht\u00eb di\u00e7ka si sistemi i loj\u00ebrave Atari, i krijuar nga DeepMind, n\u00eb t\u00eb cilin ai kishte t\u00eb kujtuar se \u00e7far\u00eb t\u00eb b\u00ebnte kur shihte piksel n\u00eb vende t\u00eb caktuara n\u00eb ekran. N\u00ebse merrni mjaft t\u00eb dh\u00ebna, mund t\u00eb duket se keni kuptim, por n\u00eb t\u00eb v\u00ebrtet\u00eb \u00ebsht\u00eb shum\u00eb sip\u00ebrfaq\u00ebsor. Nj\u00eb d\u00ebshmi p\u00ebr k\u00ebt\u00eb \u00ebsht\u00eb se n\u00ebse l\u00ebvizni objekteve p\u00ebr tre piksel, AI-pika luan shum\u00eb m\u00eb keq. Ndryshimet e v\u00ebn\u00eb at\u00eb n\u00eb nj\u00eb pozicion t\u00eb v\u00ebshtir\u00eb. Kjo \u00ebsht\u00eb e kund\u00ebrta e kuptimit t\u00eb thell\u00eb.<\/p>\n

P\u00ebr t\u00eb zgjidhur k\u00ebt\u00eb problem, ju sugjeroni t\u00eb kthehemi n\u00eb AI-n\u00eb klasike. \u00c7far\u00eb avantazhe na nevojiten t\u00eb p\u00ebrfitojm\u00eb?<\/b><\/p>\n

Ka disa avantazhe. <\/p>\n

S\u00eb pari, AI-ja klasike n\u00eb t\u00eb v\u00ebrtet\u00eb \u00ebsht\u00eb nj\u00eb korniz\u00eb p\u00ebr krijimin e modeleve kognitive t\u00eb bot\u00ebs, mbi t\u00eb cilat m\u00eb pas mund t\u00eb b\u00ebhen p\u00ebrfundime.<\/p>\n

S\u00eb dyti, AI-ja klasike \u00ebsht\u00eb n\u00eb p\u00ebrputhje t\u00eb p\u00ebrkryer me rregullat. Tani n\u00eb fush\u00ebn e m\u00ebsimit t\u00eb thell\u00eb v\u00ebrehet nj\u00eb tendenc\u00eb e \u00e7uditshme, kur specialist\u00ebt p\u00ebrpiqen t\u00eb shmangen nga rregullat. Ata duan t\u00eb b\u00ebjn\u00eb gjith\u00e7ka me rrjetet nervore dhe t\u00eb mos b\u00ebjn\u00eb asgj\u00eb q\u00eb duket si programim klasik. Por ka detyra q\u00eb jan\u00eb zgjidhur qet\u00ebsisht n\u00eb k\u00ebt\u00eb m\u00ebnyr\u00eb, dhe askush nuk e ka v\u00ebn\u00eb re. P\u00ebr shembull, nd\u00ebrtimi i itinerareve n\u00eb Google Maps.<\/p>\n

N\u00eb t\u00eb v\u00ebrtet\u00eb, na nevojiten t\u00eb dy qasjet. M\u00ebsimi i makineris\u00eb lejon t\u00eb m\u00ebsojm\u00eb mir\u00eb nga t\u00eb dh\u00ebnat, por ndihmon shum\u00eb pak n\u00eb paraqitjen e abstraksionit q\u00eb p\u00ebrfaq\u00ebson nj\u00eb program kompjuterik. Inteligjenca artificiale klasike funksionon mir\u00eb me abstraksionet, por duhet t\u00eb programohet krejt\u00ebsisht me dor\u00eb, dhe n\u00eb bot\u00eb ka grumbulluar shum\u00eb njohuri q\u00eb nuk mund t\u00eb programohen t\u00eb gjitha. \u00cbsht\u00eb e qart\u00eb se na nevojitet nj\u00eb bashkim i t\u00eb dyja qasjeve.<\/p>\n

Kjo lidhet me kapitullin ku flisni p\u00ebr at\u00eb \u00e7far\u00eb mund t\u00eb m\u00ebsojm\u00eb nga mendja njer\u00ebzore. Dhe n\u00eb radh\u00eb t\u00eb par\u00eb mbi konceptin e bazuar n\u00eb iden\u00eb e p\u00ebrmendur m\u00eb sip\u00ebr, q\u00eb nd\u00ebrgjegjja jon\u00eb p\u00ebrb\u00ebhet nga shum\u00eb sisteme diferite q\u00eb funksionojn\u00eb ndryshe.<\/b><\/p>\n

Mendoj se ka nj\u00eb tjet\u00ebr m\u00ebnyr\u00eb p\u00ebr ta shpjeguar k\u00ebt\u00eb: \u00e7do sistem kognitiv q\u00eb kemi v\u00ebrtet zgjidh detyra t\u00eb ndryshme. Pjes\u00ebt p\u00ebrkat\u00ebse t\u00eb inteligjenc\u00ebs artificiale duhet t\u00eb projektohen p\u00ebr t\u00eb zgjidhur detyra t\u00eb ndryshme, t\u00eb cilat kan\u00eb karakteristika t\u00eb ndryshme.<\/p>\n

Aktualisht ne po p\u00ebrpiqemi t\u00eb p\u00ebrdorim disa teknologji t\u00eb gjith\u00e7kaje n\u00eb nj\u00eb p\u00ebr t\u00eb zgjidhur detyra q\u00eb ndryshojn\u00eb radikalisht nga nj\u00ebra-tjetra. T\u00eb kuptuarit e nj\u00eb fraze nuk \u00ebsht\u00eb e nj\u00ebjt\u00eb me njohjen e nj\u00eb objekti. Por njer\u00ebzit p\u00ebrpiqen t\u00eb p\u00ebrdorin m\u00ebsimin e thell\u00eb n\u00eb t\u00eb dy rastet. Nga pik\u00ebpamja kognitive, k\u00ebto jan\u00eb detyra t\u00eb cil\u00ebsore t\u00eb ndryshme. Jam i habitur nga sa pak komuniteti i specialist\u00ebve t\u00eb m\u00ebsimit t\u00eb thell\u00eb vler\u00ebson inteligjenc\u00ebn artificiale klasike. Pse t\u00eb presim p\u00ebr nj\u00eb arm\u00eb argjendi? Ajo \u00ebsht\u00eb e pamundur, dhe k\u00ebrkimet e pasuksesshme nuk lejojn\u00eb t\u00eb kuptohet e gjith\u00eb kompleksiteti i detyr\u00ebs s\u00eb krijimit t\u00eb inteligjenc\u00ebs artificiale.<\/p>\n

Ju gjithashtu p\u00ebrmendni se sistemet e inteligjenc\u00ebs artificiale jan\u00eb t\u00eb nevojshme p\u00ebr t\u00eb kuptuar lidhjet shkak\u00ebsore. A mendoni se k\u00ebtu do t\u00eb na ndihmoj\u00eb m\u00ebsimi i thell\u00eb, inteligjenca artificiale klasike, apo di\u00e7ka krejt\u00ebsisht e re?<\/b><\/p>\n

Kjo \u00ebsht\u00eb nj\u00eb tjet\u00ebr fush\u00eb p\u00ebr t\u00eb cil\u00ebn m\u00ebsimi i thell\u00eb nuk \u00ebsht\u00eb aq i p\u00ebrshtatsh\u00ebm. Ai nuk shpjegon arsyet e disa ngjarjeve, por llogarit probabilitetin e nj\u00eb ngjarjeje n\u00eb kushte t\u00eb caktuara.<\/p>\n

P\u00ebr \u00e7far\u00eb po flasim? Ju shikoni disa skenar\u00eb dhe kuptoni se pse ndodhin ato dhe \u00e7far\u00eb mund t\u00eb ndodh\u00eb n\u00ebse disa rrethana ndryshojn\u00eb. Mund t\u00eb shikoj nj\u00eb mb\u00ebshtetje, mbi t\u00eb cil\u00ebn q\u00ebndron televizori, dhe t\u00eb imagjinoj se n\u00ebse ia pres nj\u00ebr\u00ebn k\u00ebmb\u00eb, mb\u00ebshtetja do t\u00eb p\u00ebrmbyset dhe televizori do t\u00eb bjer\u00eb. Kjo \u00ebsht\u00eb nj\u00eb lidhje shkak\u00ebsore.<\/p>\n

Inteligjenca klasike na ofron disa mjete p\u00ebr k\u00ebt\u00eb. Ajo mund t\u00eb tregoj\u00eb, p\u00ebr shembull, se \u00e7far\u00eb \u00ebsht\u00eb mb\u00ebshtetje dhe \u00e7far\u00eb \u00ebsht\u00eb r\u00ebnia. Por nuk do ta ekzagjeroj. Problemi \u00ebsht\u00eb se inteligjenca klasike n\u00eb shumic\u00ebn e rasteve varet nga plot\u00ebsia e informacionit p\u00ebr at\u00eb q\u00eb ndodh, nd\u00ebrsa un\u00eb arrita n\u00eb p\u00ebrfundimin vet\u00ebm duke e par\u00eb mb\u00ebshtetjen. N\u00eb nj\u00eb far\u00eb m\u00ebnyre, un\u00eb mund t\u00eb p\u00ebrmbledh, t\u00eb paraqes pjes\u00eb t\u00eb mb\u00ebshtetjes q\u00eb nuk i shoh. Ne ende nuk kemi mjete p\u00ebr t\u00eb realizuar k\u00ebt\u00eb ve\u00e7ori.<\/p>\n

Ju tregoni gjithashtu se njer\u00ebzit kan\u00eb nj\u00eb njohuri t\u00eb lindur. Si mund t\u00eb realizohet kjo n\u00eb inteligjenc\u00ebn artificiale?<\/b><\/p>\n

N\u00eb momentin e lindjes, truri yn\u00eb \u00ebsht\u00eb tashm\u00eb nj\u00eb sistem i menduar me shum\u00eb kujdes. Ai nuk \u00ebsht\u00eb i fiksuar; natyra krijoi nj\u00eb skic\u00eb t\u00eb par\u00eb, t\u00eb pap\u00ebrpunuar. M\u00eb pas, t\u00eb m\u00ebsuarit na ndihmon ta rishikojm\u00eb k\u00ebt\u00eb skic\u00eb gjat\u00eb gjith\u00eb jet\u00ebs son\u00eb.<\/p>\n

Skica e pap\u00ebrfunduar e trurit ka tashm\u00eb disa mund\u00ebsi. Nj\u00eb dele male e sapolindur ka aft\u00ebsin\u00eb p\u00ebr t\u00eb zbritur pa gabime nga mali brenda disa or\u00ebsh. \u00cbsht\u00eb e qart\u00eb se ajo tashm\u00eb ka nj\u00eb kuptim t\u00eb hap\u00ebsir\u00ebs tridimensionale, trupit t\u00eb saj dhe marr\u00ebdh\u00ebnies mes tyre. Nj\u00eb sistem t\u00ebr\u00ebsisht kompleks.<\/p>\n

P\u00ebr k\u00ebt\u00eb arsye, mendoj se na duhen hibrid. \u00cbsht\u00eb e v\u00ebshtir\u00eb t\u00eb imagjinosh se si mund t\u00eb krijosh nj\u00eb robot q\u00eb funksionon mir\u00eb n\u00eb bot\u00eb pa njohuri t\u00eb ngjashme, e ne nuk dim\u00eb nga nga duhet t\u00eb fillojm\u00eb, n\u00eb vend q\u00eb t\u00eb fillojm\u00eb nga nj\u00eb flet\u00eb e bardh\u00eb dhe t\u00eb m\u00ebsojm\u00eb nga nj\u00eb p\u00ebrvoj\u00eb t\u00eb gjat\u00eb dhe t\u00eb madhe.<\/p>\n

Sa i p\u00ebrket njer\u00ebzve, njohuria jon\u00eb e lindur vjen nga genomet tona, t\u00eb cilat jan\u00eb evoluar p\u00ebr nj\u00eb periudh\u00eb t\u00eb gjat\u00eb. Nd\u00ebrsa p\u00ebr sistemet e inteligjenc\u00ebs artificiale, do t\u00eb na duhet t\u00eb ndjekim nj\u00eb rrug\u00eb tjet\u00ebr. Pjes\u00ebrisht, k\u00ebto mund t\u00eb jen\u00eb rregulla t\u00eb nd\u00ebrtimit t\u00eb algoritmeve tona. Pjes\u00ebrisht, ato mund t\u00eb jen\u00eb rregulla p\u00ebr krijimin e strukturave t\u00eb t\u00eb dh\u00ebnave me t\u00eb cilat manipulojn\u00eb k\u00ebto algoritme. Dhe pjes\u00ebrisht, ato mund t\u00eb jen\u00eb njohuri t\u00eb cilat ne do t'i futim drejtp\u00ebrdrejt n\u00eb makina.<\/p>\n

\u00cbsht\u00eb interesante se n\u00eb lib\u00ebr, ju arrini n\u00eb iden\u00eb e besimit dhe krijimit t\u00eb sistemeve me besim. Pse e zgjodh\u00ebt pik\u00ebrisht k\u00ebt\u00eb kriter?<\/b><\/p>\n

Un\u00eb mendoj se gjith\u00e7ka sot p\u00ebrfaq\u00ebson nj\u00eb loj\u00eb topi. M\u00eb duket se po jetojm\u00eb nj\u00eb moment t\u00eb \u00e7uditsh\u00ebm n\u00eb histori, duke i besuar softuer\u00ebve q\u00eb nuk meritojn\u00eb besim. Mendoj se shqet\u00ebsimet tona t\u00eb sotme nuk do t\u00eb jen\u00eb t\u00eb p\u00ebrjetshme. Pas nj\u00ebqind vjet\u00ebsh, inteligjenca artificiale do t\u00eb justifikoj\u00eb besimin ton\u00eb, ndoshta edhe m\u00eb shpejt.<\/p>\n

Por sot, AI \u00ebsht\u00eb e rrezikshme. Jo n\u00eb kuptimin q\u00eb e frik\u00ebson Elon Musk, por n\u00eb at\u00eb q\u00eb sistemet e intervistave p\u00ebr pun\u00ebsim jan\u00eb diskriminuese ndaj grave, pavar\u00ebsisht nga ajo q\u00eb b\u00ebjn\u00eb programuesit, sepse mjetet e tyre jan\u00eb shum\u00eb t\u00eb thjeshta.<\/p>\n

Do t\u00eb doja q\u00eb t\u00eb kishim nj\u00eb AI m\u00eb t\u00eb mir\u00eb. Nuk dua q\u00eb t\u00eb filloj\u00eb 'dimri i inteligjenc\u00ebs artificiale', kur njer\u00ebzit e kuptojn\u00eb se AI nuk funksionon dhe \u00ebsht\u00eb thjesht e rrezikshme, dhe nuk do t\u00eb duan ta rregullojn\u00eb at\u00eb.<\/p>\n

N\u00eb nj\u00ebfar\u00eb m\u00ebnyre, libri juaj v\u00ebrtet duket shum\u00eb optimist. Ju supozoni se \u00ebsht\u00eb e mundur t\u00eb nd\u00ebrtohet nj\u00eb AI e besueshme. Thjesht na duhet t\u00eb shikojm\u00eb n\u00eb nj\u00eb drejtim tjet\u00ebr.<\/b><\/p>\n

E v\u00ebrtet\u00eb, libri \u00ebsht\u00eb shum\u00eb pesimist p\u00ebr perspektiv\u00ebn afatshkurt\u00ebr dhe shum\u00eb optimist p\u00ebr at\u00eb afatgjat\u00eb. Ne besojm\u00eb se t\u00eb gjitha problemet q\u00eb kemi p\u00ebrshkruar mund t\u00eb zgjidhen, n\u00ebse shikojm\u00eb n\u00eb nj\u00eb m\u00ebnyr\u00eb m\u00eb t\u00eb gjer\u00eb p\u00ebr at\u00eb se si duhet t\u00eb jen\u00eb p\u00ebrgjigjet e duhura. Dhe mendojm\u00eb se n\u00ebse ndodh kjo, bota do t\u00eb b\u00ebhet m\u00eb e mir\u00eb.
\n
Burimi: habr.com<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"

\u042d\u0442\u043e\u0442 \u0442\u0435\u043a\u0441\u0442 \u2014 \u043d\u0435 \u0440\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442 \u043d\u0430\u0443\u0447\u043d\u043e\u0433\u043e \u0438\u0441\u0441\u043b\u0435\u0434\u043e\u0432\u0430\u043d\u0438\u044f, \u0430 \u043e\u0434\u043d\u043e \u0438\u0437 \u043c\u043d\u043e\u0433\u0438\u0445 \u043c\u043d\u0435\u043d\u0438\u0439 \u043e\u0442\u043d\u043e\u0441\u0438\u0442\u0435\u043b\u044c\u043d\u043e \u043d\u0430\u0448\u0435\u0433\u043e \u0431\u043b\u0438\u0436\u0430\u0439\u0448\u0435\u0433\u043e \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0447\u0435\u0441\u043a\u043e\u0433\u043e \u0440\u0430\u0437\u0432\u0438\u0442\u0438\u044f. \u0418 \u0437\u0430\u043e\u0434\u043d\u043e \u043f\u0440\u0438\u0433\u043b\u0430\u0448\u0435\u043d\u0438\u0435 \u043a \u0434\u0438\u0441\u043a\u0443\u0441\u0441\u0438\u0438. \u0413\u0430\u0440\u0438 \u041c\u0430\u0440\u043a\u0443\u0441, \u043f\u0440\u043e\u0444\u0435\u0441\u0441\u043e\u0440 \u041d\u044c\u044e-\u0419\u043e\u0440\u043a\u0441\u043a\u043e\u0433\u043e \u0443\u043d\u0438\u0432\u0435\u0440\u0441\u0438\u0442\u0435\u0442\u0430, \u0443\u0432\u0435\u0440\u0435\u043d, \u0447\u0442\u043e \u0433\u043b\u0443\u0431\u043e\u043a\u043e\u0435 \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u0435 \u0438\u0433\u0440\u0430\u0435\u0442 \u0432\u0430\u0436\u043d\u0443\u044e \u0440\u043e\u043b\u044c \u0432 \u0440\u0430\u0437\u0432\u0438\u0442\u0438\u0438 \u0418\u0418. \u041d\u043e \u043e\u043d \u0442\u0430\u043a\u0436\u0435 \u0441\u0447\u0438\u0442\u0430\u0435\u0442, \u0447\u0442\u043e \u0438\u0437\u0431\u044b\u0442\u043e\u0447\u043d\u043e\u0435 \u0443\u0432\u043b\u0435\u0447\u0435\u043d\u0438\u0435 \u044d\u0442\u043e\u0439 \u043c\u0435\u0442\u043e\u0434\u0438\u043a\u043e\u0439 \u043c\u043e\u0436\u0435\u0442 \u043f\u0440\u0438\u0432\u0435\u0441\u0442\u0438 \u043a \u0435\u0451 \u0434\u0438\u0441\u043a\u0440\u0435\u0434\u0438\u0442\u0430\u0446\u0438\u0438. \u0412 \u0441\u0432\u043e\u0435\u0439 \u043a\u043d\u0438\u0433\u0435 Rebooting […]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":29302,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[702],"tags":[],"class_list":["post-39055","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\n","aioseo_head_json":{"title":"\ud83e\udd47Nuk mund t\u00eb besojm\u00eb sistemet AI t\u00eb nd\u00ebrtuara vet\u00ebm mbi m\u00ebsimin e thell\u00eb | ProHoster","description":"Ky tekst.","canonical_url":"https:\/\/prohoster.info\/sq\/blog\/news\/my-ne-mozhem-doveryat-ii-sistemam-postroennym-na-odnom-lish-glubokom-obuchenii","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"sq_AL","og:site_name":"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b","og:type":"article","og:title":"\ud83e\udd47\u041c\u044b \u043d\u0435 \u043c\u043e\u0436\u0435\u043c \u0434\u043e\u0432\u0435\u0440\u044f\u0442\u044c \u0418\u0418-\u0441\u0438\u0441\u0442\u0435\u043c\u0430\u043c, \u043f\u043e\u0441\u0442\u0440\u043e\u0435\u043d\u043d\u044b\u043c \u043d\u0430 \u043e\u0434\u043d\u043e\u043c \u043b\u0438\u0448\u044c \u0433\u043b\u0443\u0431\u043e\u043a\u043e\u043c \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u0438 | ProHoster","og:description":"\u042d\u0442\u043e\u0442 \u0442\u0435\u043a\u0441\u0442.","og:url":"https:\/\/prohoster.info\/sq\/blog\/news\/my-ne-mozhem-doveryat-ii-sistemam-postroennym-na-odnom-lish-glubokom-obuchenii","og:image":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:secure_url":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:width":350,"og:image:height":350,"article:published_time":"2019-10-31T19:27:36+00:00","article:modified_time":"2019-10-31T19:27:36+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"39055","title":null,"description":null,"keywords":null,"keyphrases":null,"primary_term":null,"canonical_url":null,"og_title":null,"og_description":null,"og_object_type":"default","og_image_type":"default","og_image_url":null,"og_image_width":null,"og_image_height":null,"og_image_custom_url":null,"og_image_custom_fields":null,"og_video":null,"og_custom_url":null,"og_article_section":null,"og_article_tags":null,"twitter_use_og":false,"twitter_card":"default","twitter_image_type":"default","twitter_image_url":null,"twitter_image_custom_url":null,"twitter_image_custom_fields":null,"twitter_title":null,"twitter_description":null,"schema":{"blockGraphs":[],"customGraphs":[],"default":{"data":{"Article":[],"Course":[],"Dataset":[],"FAQPage":[],"Movie":[],"Person":[],"Product":[],"ProductReview":[],"Car":[],"Recipe":[],"Service":[],"SoftwareApplication":[],"WebPage":[]},"graphName":"","isEnabled":true},"graphs":[]},"schema_type":null,"schema_type_options":null,"pillar_content":false,"robots_default":true,"robots_noindex":false,"robots_noarchive":false,"robots_nosnippet":false,"robots_nofollow":false,"robots_noimageindex":false,"robots_noodp":false,"robots_notranslate":false,"robots_max_snippet":null,"robots_max_videopreview":null,"robots_max_imagepreview":"large","priority":null,"frequency":null,"local_seo":null,"seo_analyzer_scan_date":"2026-01-24 00:33:19","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2021-03-01 00:57:31","updated":"2026-01-24 00:33:19","focus_keyword":null,"additional_keywords":null,"truseo_locale":null},"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/posts\/39055","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/comments?post=39055"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/posts\/39055\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/media\/29302"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/media?parent=39055"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/categories?post=39055"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/tags?post=39055"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}