ব্যবহারকারী:Al-mumtahinah/কৃত্রিম বুদ্ধিমত্তা

চিত্র:TOPIO 3.0.jpg
TOPIO, a humanoid robot, played table tennis at Tokyo International Robot Exhibition (IREX) 2009.[]

কৃত্রিম বুদ্ধিমত্তা (AI) বলতে বোঝায় যন্ত্রের বুদ্ধিমত্তাকে এবং কম্পিউটার বিজ্ঞানের সেই শাখাকে যা মানবনির্মিত যন্ত্রের মধ্যে বুদ্ধিমত্তাকে সৃষ্টি করতে ইচ্ছুক। কৃত্রিম বুদ্ধিমত্তার পাঠ্যপুস্তকসমূহে এই ক্ষেত্রটিকে বর্ণণা করা হয়েছে এই বলে যে, এ হলো 'বুদ্ধিমান' কিছু[] নির্মাণ ও তার জন্য অধ্যয়নের এক ক্ষেত্র। এখানে বুদ্ধিমান কিছু বলতে বোঝাচ্ছে এমন এক 'তন্ত্'র যা তার পারিপার্শ্বিক পরিবেশকে অনুধাবন ও মূল্যায়ন করতে সক্ষম এবং তদানুসারে এমন সিদ্ধান্ত গ্রহণে সক্ষম যা তার কার্যকারীতার সম্ভাবনাকে উন্নীত করবে। []কৃত্রিম বুদ্ধিমত্তার ইংরাজী নামের কথাটি, অর্থাত্ Artificial Intelligence কথাটি ১৯৫৬ সালে প্রথম চালু করেন[]জন ম্যাককার্থি। তিনি কৃত্রিম বুদ্ধিমত্তার সংজ্ঞার্থ দিয়েছেন এই ভাবে যে, 'কৃত্রিম বুদ্ধিমত্তা হল বুদ্ধিমান যন্ত্র নির্মাণের বিজ্ঞান ও প্রযুক্তি'। []

বিজ্ঞানের এই ক্ষেত্রটি স্থাপিত হয়েছিল এই দাবীর উপর ভিত্তি করে যে, বুদ্ধিমত্তা, যা কিনা মানুষের এক কেন্দ্রীয় বৈশিষ্ট্য, --Homo sapiens এর sapience--সেই বুদ্ধিমত্তাকে এতই সুচারু ভাবে বর্ণনা করা সম্ভব যে তাকে কোন যন্ত্রের মধ্যেও অুনকৃত করা যায়। []এর ফলে মনেরপ্রকৃতি কি সে বিষয়ে এবং কৃত্রিম ভাবে প্রস্তুত সত্তার নির্মাণের নৈতিকতা নিয়েও দার্শনিক প্রশ্ন উঠে গেল, যে সব বিষয় গুলিকে উপকথা, গল্প এবং দর্শন since antiquity এর মধ্যে নির্দেশিত করা হয়েছে।[] কৃত্রিম বুদ্ধিমত্তা যেন আশাবাদের বিষয় হয়ে গেছে।[] কিন্তু সেই সঙ্গে এর আবার বাধা বিপত্তির ব্যাপারও আছে।[]আর আজ তো এটা প্রযুক্তি শিল্পের একটা অতি প্রয়োজনীয় অংশই হয়ে দাঁড়িয়েছে, যা কম্পিউটার বিজ্ঞানের বহু দুরূহ সমস্যার তুমুল সমাধান দিয়ে যাচ্ছে।[১০] কৃত্রিম বুদ্ধিমত্তার গবেষণার ক্ষেত্রটি কিন্তু অতিমাত্রায় বিশেযায়িত, গভীর ভাবে নানা অধীনস্ত ক্ষেত্রে বিভক্ত, অবস্থা এমনই যে তাদের নিজেদের মধ্যে অনেক সময়েই পারস্পরিক যোগাযোগের দায়টা ঠিকমত পালন করতে পারে না। [১১] ঐ সব অধীন ক্ষেত্রগুলি বিশেষ বিশেষ সংস্থাকে ঘিরে বড় হয়েছে, কখনো একক গবেষকদের কাজের মাধ্যমে, কখনো বিশেষ কোন সমস্যার সমাধান করতে গিয়ে, কখনো বা কি ভাবে কৃত্রিম বুদ্ধিমত্তার কাজ করা চলবে এ নিয়ে দীর্ঘসূত্রী মতপার্থক্যে আর ভিন্ন ভিন্ন বিশাল তফাত্ যুক্ত উপায় সমূহের প্রয়োগ কৌশলের ফলে। কৃত্রিম বুদ্ধিমত্তার কেন্দ্রীয় সমস্যার মধ্যে আছে নানা ধরণের বৈশিষ্ট্য. যেমন যুক্তি প্রয়োগ, জ্ঞান, পরিকল্পনা, শিখন, পারস্পরিক যোগাযোগ, সংবেদন এবং দ্রব্যাদিকে নড়ানো-সরানো বা তাদের কাজে লাগানো। [১২] সাধারণ বুদ্ধিমত্তা(বা স্ট্রং এ আই) কিন্তু এখনো এই ক্ষেত্রের সুদূর লক্ষ্য সমূহের মধ্যে অন্যতম।[১৩]

ইতিহাস

সম্পাদনা

চিন্তাশীল যন্ত্র আর কৃত্রিম সত্তার দেখা পাওয়া গেছে গ্রীক উপকথা সমূহে, যেমন ক্রীট এর তালোস, হেফিস্তাসএর ব্রোঞ্জ রোবো, এবং পিগম্যালিয়নের গালাতিয়া[১৪]প্রতিটি প্রধান সভ্যতাতেই মনবাকৃতির এমন কিছু কিছু নির্মাণ করা হয়েছে যাদেরকে বিশ্বাস করা হয়েছে বুদ্ধিমত্তা আছে বলে। সঞ্চালনক্ষম cult imageসমূহের পূজো করা হত মিশরআর গ্রীসএ[১৫] এবং মানবক (humanoid)বানিয়েছিলেন ইয়ান শি, আলেকজান্দ্রিয়ার বীর এবং আল-জাজ়ারি[১৬]এমন ব্যাপক বিশ্বাসও চালু আছে যে জা-বির ইবন্ হাইয়ান, জুডা লৌইপারাক্লিসাস পর্যন্তও কৃত্রিম সত্তা তৈরি করেছিলেন।[১৭] উনবিংশ ও বিংশ শতকে তো গল্প সাহিত্যে কৃত্রিম সত্তা একটা খুবই সাধারণ ব্যাপার হয়ে দাঁড়াল। যেমন, মেরী শেলীর ফ্র্যাংকেনস্টাইন বা কারেল চাপেকএর R.U.R. (Rossum's Universal Robots)[১৮]পামেলা ম্যাককর্ডাকএর বিশ্লেষণ হল এই যে , এই সব উদাহরণ সমূহ হল (পামেলার নিজের ভাষায়) "to forge the gods"এর এক সুপ্রাচীন তাড়নার নিদর্শন।[]এই সব সত্তার গল্প সমূহ আর তাদের পরিণতির মধ্যে আলোচিত হয়েছে একই ধরণের আশা, আশঙ্কা আর নৈতিক উদ্বেগ সমূহ যা কৃত্রিম বুদ্ধিমত্তার দ্বারা উপস্থাপিত।

দার্শনিকগণ এবং গণিতজ্ঞেরা সুপ্রচীন কাল থেকেই যান্ত্রিক অথবা "আকারগত" যুক্তিকৌশল গড়ে তুলেছেন । অ্যালান টুরিং এবং আরো অনেক গণিতজ্ঞের কাজের উপর ভিত্তি করে তর্কবিদ্যার অধ্যয়ন তো সরাসরিই নিয়ে গেছে প্রোগ্রামেবল ডিজিট্যাল ইলেকট্রনিক কম্পিউটারএর আবিষ্কারের মধ্যে। টুরিং এর থিওরি অব কম্পিউটেশন এ প্রস্তাব করা হয়েছিল যে, কোন যন্ত্র, স্রেফ ০ আর ১ এর মত সরল প্রতীক সমূহ ঘেঁটেই যে কোন বুঝতে পারার মত গাণিতিক উপপাদ্যকে অনুকৃত করতে সক্ষম। [১৯][২০] স্নায়ুবিদ্যা, ইনফর্মেশন থিওরি and সাইবারনেটিক্সএর ক্ষেত্র সমূহে সমসাময়িক সহগামী আবিষ্কার গুলি সহ এটাই , গবেষকদের একটা ছোট দলকে অনুপ্রাণিত করল ইলেকট্রনিক মগজ তৈরির সম্ভাব্যতার বিষয়টি নিয়ে গুরতরভাবে বিবেচনা করা শুরু করতে।[২১]


১৯৫৬ সালের গ্রীষ্মে ডার্টমাউথ কলেজ ক্যাম্পাসের এক সম্মেলনে কৃত্রিম বুদ্ধিমত্তার গবেষণার ক্ষেত্রটি প্রতিষ্ঠিত হয়। উপস্থিত ব্যক্তিবর্গ,যাঁদের মধ্যে ছিলেন জন ম্যাককার্থি, মার্ভিন মিনস্কি, অ্যালেন নিউওয়েল আর হার্বার্ট সাইমন এঁরাও, এঁরাই পেয়ে গেলেন পরের বেশ কয়েক দশকের জন্য কৃত্রিম বুদ্ধিমত্তার গবেষণার ক্ষেত্রের নেতৃত্ব। [২২] তাঁরা ও তাঁদের ছাত্ররা এমন সব প্রোগ্রাম লিখলেন যেগুলো, বেশির ভাগ মানুষের কাছেই, স্রেফ অদ্ভুত ব্যাপার:[২৩] কম্পিউটারেরা বীজগণিতে ভাষায় প্রকাশিত সমস্যার সমাধান করে দিচ্ছে, যৌক্তিক উপপাদ্য প্রমাণ করে দিচ্ছে, এমনকি ইংরাজীতে কথাও বলছে।[২৪]১৯৬০ এর দশকের মাঝামাঝি নাগাদ,মার্কিন যুক্তরাষ্ট্রে গবেষণা চলতে লাগল ডিপার্টমেন্ট অব ডিফেন্স এর মোটা অনুদানে [২৫] আর গোটা পৃথিবী জুড়েই গবেষণাগার স্থাপিত হতে লাগল।[২৬]কৃত্রিম বুদ্ধমত্তা ক্ষেত্রটির স্থাপিয়তারা এই নতুন ক্ষেত্রের ভবিষ্যত্ বিষয়ে সুগভীর ভাবে আশাবাদী ছিলেন: হার্বার্ট সাইমন ভবিষ্যদ্বাণী করেছিলেন যে, "বিশ বছরের ভিতরেই , এক জন মানুষ যা যা করতে সক্ষম তার সবই যন্ত্রেরা করতে সক্ষম হয়ে যাবে" আর মার্ভিন মিনস্কিও একমত হয়েছিলেন এই লিখে যে, "এক প্রজন্মের ভিতরেই ... 'কৃত্রিম বুদ্ধিমত্তা' বানানোর সমস্যাটি প্রায় সবটাই সমাধা হয়ে যাবে।"[২৭]


দুর্ভাগ্যবশতঃ, যে সব সমস্যার সম্মুখীন তাঁরা হয়েছিলেন তাদের কয়েকটির স্বরূপ তাঁরা উপলব্ধি করতে ব্যর্থ হয়েছিলেন।[২৮] ১৯৭৪ সালে, স্যার জেমস লাইটহিলএর এক সমালোচনার প্রতিক্রিয়ায় আর মার্কিন যুক্তরাষ্ট্রের কংগ্রেসের কাছ থেকে আরও বেশী কাজের কাজে টাকা ঢালবার জন্য আসতে থাকা ক্রমাগত চাপের মুখে পড়ে মার্কিন যুক্তরাষ্ট্রের এবং ব্রিটিশ যুক্তরাজ্যের উভয় সরকারই AI ক্ষেত্রে অনির্দিষ্ট লক্ষ্যাভিমুখী আর কেবলমাত্র রহস্য উন্মোচনাত্মক সমস্ত রকম গবেষণাতে রসদ দেওয়া থেকে হাত গুটিয়ে ফেলল। পরবর্তী কয়েকটা বছরকে, যে সময়টায় ঐ সব প্রকল্প গুলোর জন্য টাকা পাওয়াই মুশকিলের ব্যাপার হয়ে গেল, সেই সময়টাকে অনেক পরে বলা হবে "AI winter" অর্থাত্ কৃত্রিম বুদ্ধিমত্তার শীতের দিন (আসলে শীতের দেশে শীতের সময়টা খুবই কঠিন সময় কিনা!)[২৯]

চাকা ঘুরে গেল ১৯৮০ এর দশকের গোড়ার দিকে। এসে গেল এক্সপার্ট সিস্টেম বা বিশেষজ্ঞ তন্ত্র। তুমুল ব্যবসায়িক সাফল্য পেল সেটা। আর তারই জোয়ার লেগে AI এর গবেষণা পুনরুজ্জীবিত হয়ে উঠল।[৩০] এই এক্সপার্ট সিস্টেম ব্যাপারটা আসলে কিছু না, এক ধরণের AI প্রোগ্রাম, যা এক বা একাধিক বিশেষজ্ঞ মানুষের জ্ঞান আর দক্ষতার ভাণ্ডারকে নকল করতে পারে। (যেন যান্ত্রিক ডাক্তার, উকিল, শিক্ষক, এইসব আর কি।) ১৯৮৫ সাল নাগাদ AIএর বাজার এক শ কোটি ডলার ছাড়িয়ে গেল। একই সঙ্গে, জাপানের পঞ্চম প্রজন্মের কম্পিউটার প্রকল্প সমূহ মার্কিন এবং ব্রিটিশ সরকারকে অনুপ্রাণিত করল ঐ ক্ষেত্রে শিক্ষাগত গবেষণায় পুনরায় টাকা ঢালতে। [৩১] সে হলে কি হয়, ফের ১৯৮৭ সালে লিস্প মেশিনএর বাজারের ভেঙে পড়া দিয়ে শুরু হয়ে AIএর ক্ষেত্রটি পুনরায় দুর্নামের মধ্যে পড়ল। শুরু হল দ্বিতীয় আর এক, দীর্ঘতরভাবে স্থায়ী AI শীত[৩২]


১৯৯০ এর দশক জুড়ে আর একুশ শতকের গোড়ার দিকে, AI লাভ করল তার সর্বোচ্চ সাফল্য, যদিও তা কিছুটা পর্দার আড়ালেই থেকে গেল। কৃত্রিম বুদ্ধিমত্তার ব্যবহার হতে লাগল সরবরাহ, তথ্য অনুসন্ধান, চিকিত্সার মত ক্ষেত্রে। এছাড়াও সারা প্রযুক্তি শিল্প জুড়ে আরো নানা ক্ষেত্রে।[১০] এই সাফল্যের অনেক গুলো কারণ ছিল। যেমন, কম্পিউটারের গণনা ক্ষমতার ক্রমাগত বৃদ্ধি (এবিষয়ে মূর এর সূত্র দেখুন), তেমনি, কোন বড় সমস্যার অধীন সুনির্দিষ্ট ছোট ছোট সমস্যার সমাধানের উপর অধিকতর গুরুত্ব প্রদান, এছাড়া , AI এবং অন্যান্য যেসব ক্ষেত্র একই ধরণের সমস্যার সমাধানে কাজ করছে তাদের মধ্যে নতুন নতুন বন্ধন গড়ে তোলা, তেমনি আবার গবেষকদের মধ্যেও সুদৃঢ় গাণিতিক পদ্ধতির প্রতি আর সুকঠিন বৈজ্ঞানিক মানদণ্ডের প্রতি একটা নতুন দায়বদ্ধতা জন্ম এসবও এর কারণ। [৩৩] ১১ ই মে ১৯৯৪, ডীপ ব্লু নামে এক প্রথম দাবাড়ু কম্পিউটারের উদয় হল যে তখনকার বিশ্ব চ্যাম্পিয়ান গ্যারি ক্যাস্পারভকে পরাজিত করতে সক্ষম হল।[৩৪] ২০০৫ এ, স্ট্যানফোর্ড জাত এক রোবট এল যে মরুভূমির এক না দেখা পথে স্বয়ংক্রিয় ভাবে চলে ১৩১ মাইল পরিভ্রমণ কর DARPA গ্র্যাণ্ড চ্যালেঞ্জজিতে নিল।[৩৫] এর দু বছর পর, CMU থেকে একটি দল এক শহুরে পরিবেশে স্বয়ংচালিত হয়ে ৫৫ মাইল পথ সমস্ত রকম ট্রাফিক নিয়মকানুন মেনে আর সমস্ত রকম ট্রাফিক ঝঞ্ঝাট অতিক্রম করে জিতে নিল DARPA আর্বান চ্যালেঞ্জ[৩৬] ফেব্রুয়ারী ২০১১ তে , জিওপার্ডি! ক্যুইজ় প্রদর্শনী নামের একটি একজ়িবিশন ম্যাচে, আই বি এম এর ওয়াটসান নামক প্রশ্নোত্তর যন্ত্রটি, ব্র্যাড রুটার আর কেন জেনিংস নামের দু জন সব সেরা "জিওপার্ডি!" চ্যাম্পিয়নকে শোচনীয় ভাবে পরাজিত করল [৩৭]

কৃত্রিম বুদ্ধিমত্তার গবেষণার মুখ্য সংজ্ঞাটি কিন্তু সময়ের সঙ্গে সঙ্গে পাল্টে চলেছে। এখনকার মত একটা কেজো সংজ্ঞা হল এই রকম: "AI গবেষণা হল তাই যা কম্পিউটার বিজ্ঞানীরা জানেনই না বর্তমানে কিভাবে তা যুক্তিপূর্ণ খরচের মধ্যে করা যায়।" উদাহরণ: ১৯৫৬ সালে অপটিক্যাল ক্যারেক্টার রিকগনিশন (অর্থাত্ আলোর সাহায্যে অক্ষর দেখে শনাক্তকরণ বা OCR) কে AI বলেই ধরা হত, কিন্তু আজ তো বেশির ভাগ ইমেজ স্ক্যানারএর সঙ্গেই, চমত্কার সব ও সি আর সফ্টওয়্যার বিনা মূল্যে পাওয়া যায়, যারা আবার নিজে নিজেই বানান পরীক্ষকও বটে, এমনকি ব্যাকরণের অশুদ্ধিও সংশোধনে সক্ষম । আজ আর কেউ ও সি আর ধরণের কম্পিউটারের সমস্যা যাদের সমাধান ইতোমধ্যেই হয়ে গেছে, তাদেরকে আর "কৃত্রিম বুদ্ধিমত্তা" বলে ডাকতে রাজী নয়।

ট্যাবলেট কম্পিউটারদের ভিতর স্বল্পমূল্যের মজাদার দাবাড়ু সফ্টওয়্যার পাওয়া তো এখন জলভাত। DARPA এখন আর দাবাড়ু কম্পিউটার সিস্টেম বানানোর জন্য তেমন খরচ টরচ করে না। ওদিকে কাইনেক্টসংস্থার এক্সবক্স ৩৬০ এর জন্য ত্রিমাত্রিক দেহ সঞ্চালনের জন্য যে ইন্টারফেসের ব্যবস্থা, তার ভিত্তি হল এক এমন অ্যালগরিথম যা এক সুদীর্ঘ AI গবেষণার ফলে বেরিয়ে এসেছে।[৩৮] কিন্তু এর ক জন উপভোক্তা ঐ প্রযুক্তিগত উত্সটির খোঁজ রাখেন?

AI এর প্রয়োগ সমূহ এখন আর কোন মতেই শুধু মাত্র প্রতিরক্ষা দপ্তরএর, আর অ্যান্ড ডি বিভাগের আওতাভুক্ত ব্যাপার নয়। এসব এখন মামুলি উপভোক্তা-দ্রব্যাদি আর সস্তা বুদ্ধিমান খেলনা জাতীয় ব্যাপার হয়ে দাঁড়িয়েছে।

চলতি কথায় , "AI"লফজটি দেখা যাচ্ছে যেন, সেই সব কম্পিউটার সমস্যার ব্যাপারে প্রযুক্ত হওয়ার যোগ্যতা হারিয়েছে যাদের সমাধান হয়ে গেছে, যাদের চাইলেই পাওয়া যাবে এমন উত্পাদিত দ্রব্য হিসাবে গণ্য করা চলে।অথচ ওদের গড়ে ওঠা কত বছরের সুদীর্ঘ পরিশ্রম সাধ্য AI গবেষণার মধ্য দিয়ে!










------------->


সমস্যাবলী

সম্পাদনা

"কোন যন্ত্র কি আদৌ বুদ্ধিমানের মত আচরণ করতে সক্ষম?" এই যে প্রশ্ন, এটা এখনো পর্যন্ত একটা সমাধান না হওয়া সমস্যা হয়েই রয়ে গেছে। "যন্ত্রের পক্ষে বুদ্ধিমানের মত আচরণ করা সম্ভব" এই বক্তব্যকে একটা কার্যকরী প্রকল্প হিসেবে ধরে নিয়েই বহু গবেষক এমনতর যন্ত্র নির্মাণে সচেষ্ট হয়েছেন।

বুদ্ধিমত্তাকে অনুকরণ (বা সৃষ্টি) করার আসল সমস্যাগুলোকে অনেকগুলো সুনির্দিষ্ট ছোট-ছোট টুকরো সমস্যায় ভাগ করা হয়েছে। এর মধ্যে আছে বিশেষ কতগুলো ধর্ম বা সক্ষমতা যেগুলোকে কোন বুদ্ধিমান তন্ত্র প্রদর্শন করতে পারলে গবেষকরা খুব খুশী হতেন। নীচে বর্ণিত ধর্মগুলোই সবচেয়ে বেশী মনোযোগ দাবি করে। [১২]

অবরোহী যুক্তি, যুক্তিশৃঙ্খল বা ন্যায়, সমস্যা সমাধান

সম্পাদনা

<-- ===Deduction, reasoning, problem solving ===--> Early AI researchers developed algorithms that imitated the step-by-step reasoning that humans use when they solve puzzles or make logical deductions.[৩৯] By the late 1980s and '90s, AI research had also developed highly successful methods for dealing with uncertain or incomplete information, employing concepts from probability and economics.[৪০]

For difficult problems, most of these algorithms can require enormous computational resources — most experience a "combinatorial explosion": the amount of memory or computer time required becomes astronomical when the problem goes beyond a certain size. The search for more efficient problem-solving algorithms is a high priority for AI research.[৪১]

Human beings solve most of their problems using fast, intuitive judgments rather than the conscious, step-by-step deduction that early AI research was able to model.[৪২] AI has made some progress at imitating this kind of "sub-symbolic" problem solving: embodied agent approaches emphasize the importance of sensorimotor skills to higher reasoning; neural net research attempts to simulate the structures inside human and animal brains that give rise to this skill.

Knowledge representation

সম্পাদনা
 
An ontology represents knowledge as a set of concepts within a domain and the relationships between those concepts.

Knowledge representation[৪৩] and knowledge engineering[৪৪] are central to AI research. Many of the problems machines are expected to solve will require extensive knowledge about the world. Among the things that AI needs to represent are: objects, properties, categories and relations between objects;[৪৫] situations, events, states and time;[৪৬] causes and effects;[৪৭] knowledge about knowledge (what we know about what other people know);[৪৮] and many other, less well researched domains. A representation of "what exists" is an ontology (borrowing a word from traditional philosophy), of which the most general are called upper ontologies.[৪৯]

Among the most difficult problems in knowledge representation are:

Default reasoning and the qualification problem
Many of the things people know take the form of "working assumptions." For example, if a bird comes up in conversation, people typically picture an animal that is fist sized, sings, and flies. None of these things are true about all birds. John McCarthy identified this problem in 1969[৫০] as the qualification problem: for any commonsense rule that AI researchers care to represent, there tend to be a huge number of exceptions. Almost nothing is simply true or false in the way that abstract logic requires. AI research has explored a number of solutions to this problem.[৫১]
The breadth of commonsense knowledge
The number of atomic facts that the average person knows is astronomical. Research projects that attempt to build a complete knowledge base of commonsense knowledge (e.g., Cyc) require enormous amounts of laborious ontological engineering — they must be built, by hand, one complicated concept at a time.[৫২] A major goal is to have the computer understand enough concepts to be able to learn by reading from sources like the internet, and thus be able to add to its own ontology.[তথ্যসূত্র প্রয়োজন]
The subsymbolic form of some commonsense knowledge
Much of what people know is not represented as "facts" or "statements" that they could express verbally. For example, a chess master will avoid a particular chess position because it "feels too exposed"[৫৩] or an art critic can take one look at a statue and instantly realize that it is a fake.[৫৪] These are intuitions or tendencies that are represented in the brain non-consciously and sub-symbolically.[৫৫] Knowledge like this informs, supports and provides a context for symbolic, conscious knowledge. As with the related problem of sub-symbolic reasoning, it is hoped that situated AI or computational intelligence will provide ways to represent this kind of knowledge.[৫৫]
 
A hierarchical control system is a form of control system in which a set of devices and governing software is arranged in a hierarchy.

Intelligent agents must be able to set goals and achieve them.[৫৬] They need a way to visualize the future (they must have a representation of the state of the world and be able to make predictions about how their actions will change it) and be able to make choices that maximize the utility (or "value") of the available choices.[৫৭]

In classical planning problems, the agent can assume that it is the only thing acting on the world and it can be certain what the consequences of its actions may be.[৫৮] However, if this is not true, it must periodically check if the world matches its predictions and it must change its plan as this becomes necessary, requiring the agent to reason under uncertainty.[৫৯]

Multi-agent planning uses the cooperation and competition of many agents to achieve a given goal. Emergent behavior such as this is used by evolutionary algorithms and swarm intelligence.[৬০]

Machine learning[৬১] has been central to AI research from the beginning.[৬২] In 1956, at the original Dartmouth AI summer conference, Ray Solomonoff wrote a report on unsupervised probabilistic machine learning: "An Inductive Inference Machine".[৬৩] Unsupervised learning is the ability to find patterns in a stream of input. Supervised learning includes both classification and numerical regression. Classification is used to determine what category something belongs in, after seeing a number of examples of things from several categories. Regression is the attempt to produce a function that describes the relationship between inputs and outputs and predicts how the outputs should change as the inputs change. In reinforcement learning[৬৪] the agent is rewarded for good responses and punished for bad ones. These can be analyzed in terms of decision theory, using concepts like utility. The mathematical analysis of machine learning algorithms and their performance is a branch of theoretical computer science known as computational learning theory.[৬৫]

Natural language processing

সম্পাদনা
 
A parse tree represents the syntactic structure of a sentence according to some formal grammar.

Natural language processing[৬৬] gives machines the ability to read and understand the languages that humans speak. Many researchers hope that a sufficiently powerful natural language processing system would be able to acquire knowledge on its own, by reading the existing text available over the internet. Some straightforward applications of natural language processing include information retrieval (or text mining) and machine translation.[৬৭]

Motion and manipulation

সম্পাদনা

The field of robotics[৬৮] is closely related to AI. Intelligence is required for robots to be able to handle such tasks as object manipulation[৬৯] and navigation, with sub-problems of localization (knowing where you are), mapping (learning what is around you) and motion planning (figuring out how to get there).[৭০]

Machine perception[৭১] is the ability to use input from sensors (such as cameras, microphones, sonar and others more exotic) to deduce aspects of the world. Computer vision[৭২] is the ability to analyze visual input. A few selected subproblems are speech recognition,[৭৩] facial recognition and object recognition.[৭৪]

Social intelligence

সম্পাদনা
 
Kismet, a robot with rudimentary social skills

[তথ্যসূত্র প্রয়োজন]

Emotion and social skills[৭৫] play two roles for an intelligent agent. First, it must be able to predict the actions of others, by understanding their motives and emotional states. (This involves elements of game theory, decision theory, as well as the ability to model human emotions and the perceptual skills to detect emotions.) Also, for good human-computer interaction, an intelligent machine needs to display emotions. At the very least it must appear polite and sensitive to the humans it interacts with. At best, it should have normal emotions itself.

A sub-field of AI addresses creativity both theoretically (from a philosophical and psychological perspective) and practically (via specific implementations of systems that generate outputs that can be considered creative, or systems that identify and assess creativity). Related areas of computational research are Artificial intuition and Artificial imagination.[তথ্যসূত্র প্রয়োজন]

General intelligence

সম্পাদনা

Most researchers hope that their work will eventually be incorporated into a machine with general intelligence (known as strong AI), combining all the skills above and exceeding human abilities at most or all of them.[১৩] A few believe that anthropomorphic features like artificial consciousness or an artificial brain may be required for such a project.[৭৬][৭৭]

Many of the problems above are considered AI-complete: to solve one problem, you must solve them all. For example, even a straightforward, specific task like machine translation requires that the machine follow the author's argument (reason), know what is being talked about (knowledge), and faithfully reproduce the author's intention (social intelligence). Machine translation, therefore, is believed to be AI-complete: it may require strong AI to be done as well as humans can do it.[৭৮]

There is no established unifying theory or paradigm that guides AI research. Researchers disagree about many issues.[৭৯] A few of the most long standing questions that have remained unanswered are these: should artificial intelligence simulate natural intelligence by studying psychology or neurology? Or is human biology as irrelevant to AI research as bird biology is to aeronautical engineering?[৮০] Can intelligent behavior be described using simple, elegant principles (such as logic or optimization)? Or does it necessarily require solving a large number of completely unrelated problems?[৮১] Can intelligence be reproduced using high-level symbols, similar to words and ideas? Or does it require "sub-symbolic" processing?[৮২] John Haugeland, who coined the term GOFAI (Good Old-Fashioned Artificial Intelligence), also proposed that AI should more properly be referred to as synthetic intelligence,[৮৩] a term which has since been adopted by some non-GOFAI researchers.[৮৪][৮৫]

Cybernetics and brain simulation

সম্পাদনা
 
There is currently no consensus on how closely the brain should be simulated.

In the 1940s and 1950s, a number of researchers explored the connection between neurology, information theory, and cybernetics. Some of them built machines that used electronic networks to exhibit rudimentary intelligence, such as W. Grey Walter's turtles and the Johns Hopkins Beast. Many of these researchers gathered for meetings of the Teleological Society at Princeton University and the Ratio Club in England.[২১] By 1960, this approach was largely abandoned, although elements of it would be revived in the 1980s.

When access to digital computers became possible in the middle 1950s, AI research began to explore the possibility that human intelligence could be reduced to symbol manipulation. The research was centered in three institutions: CMU, Stanford and MIT, and each one developed its own style of research. John Haugeland named these approaches to AI "good old fashioned AI" or "GOFAI".[৮৬]

Cognitive simulation
Economist Herbert Simon and Allen Newell studied human problem-solving skills and attempted to formalize them, and their work laid the foundations of the field of artificial intelligence, as well as cognitive science, operations research and management science. Their research team used the results of psychological experiments to develop programs that simulated the techniques that people used to solve problems. This tradition, centered at Carnegie Mellon University would eventually culminate in the development of the Soar architecture in the middle 80s.[৮৭][৮৮]
Logic-based
Unlike Newell and Simon, John McCarthy felt that machines did not need to simulate human thought, but should instead try to find the essence of abstract reasoning and problem solving, regardless of whether people used the same algorithms.[৮০] His laboratory at Stanford (SAIL) focused on using formal logic to solve a wide variety of problems, including knowledge representation, planning and learning.[৮৯] Logic was also focus of the work at the University of Edinburgh and elsewhere in Europe which led to the development of the programming language Prolog and the science of logic programming.[৯০]
"Anti-logic" or "scruffy"
Researchers at MIT (such as Marvin Minsky and Seymour Papert)[৯১] found that solving difficult problems in vision and natural language processing required ad-hoc solutions – they argued that there was no simple and general principle (like logic) that would capture all the aspects of intelligent behavior. Roger Schank described their "anti-logic" approaches as "scruffy" (as opposed to the "neat" paradigms at CMU and Stanford).[৮১] Commonsense knowledge bases (such as Doug Lenat's Cyc) are an example of "scruffy" AI, since they must be built by hand, one complicated concept at a time.[৯২]
Knowledge-based
When computers with large memories became available around 1970, researchers from all three traditions began to build knowledge into AI applications.[৯৩] This "knowledge revolution" led to the development and deployment of expert systems (introduced by Edward Feigenbaum), the first truly successful form of AI software.[৩০] The knowledge revolution was also driven by the realization that enormous amounts of knowledge would be required by many simple AI applications.

During the 1960s, symbolic approaches had achieved great success at simulating high-level thinking in small demonstration programs. Approaches based on cybernetics or neural networks were abandoned or pushed into the background.[৯৪] By the 1980s, however, progress in symbolic AI seemed to stall and many believed that symbolic systems would never be able to imitate all the processes of human cognition, especially perception, robotics, learning and pattern recognition. A number of researchers began to look into "sub-symbolic" approaches to specific AI problems.[৮২]

Bottom-up, embodied, situated, behavior-based or nouvelle AI
Researchers from the related field of robotics, such as Rodney Brooks, rejected symbolic AI and focused on the basic engineering problems that would allow robots to move and survive.[৯৫] Their work revived the non-symbolic viewpoint of the early cybernetics researchers of the 50s and reintroduced the use of control theory in AI. This coincided with the development of the embodied mind thesis in the related field of cognitive science: the idea that aspects of the body (such as movement, perception and visualization) are required for higher intelligence.
Computational Intelligence
Interest in neural networks and "connectionism" was revived by David Rumelhart and others in the middle 1980s.[৯৬] These and other sub-symbolic approaches, such as fuzzy systems and evolutionary computation, are now studied collectively by the emerging discipline of computational intelligence.[৯৭]

In the 1990s, AI researchers developed sophisticated mathematical tools to solve specific subproblems. These tools are truly scientific, in the sense that their results are both measurable and verifiable, and they have been responsible for many of AI's recent successes. The shared mathematical language has also permitted a high level of collaboration with more established fields (like mathematics, economics or operations research). Stuart Russell and Peter Norvig describe this movement as nothing less than a "revolution" and "the victory of the neats."[৩৩] Critiques argue that these techniques are too focussed on particular problems and have failed to address the long term goal of general intelligence.[তথ্যসূত্র প্রয়োজন]

Integrating the approaches

সম্পাদনা
Intelligent agent paradigm
An intelligent agent is a system that perceives its environment and takes actions which maximize its chances of success. The simplest intelligent agents are programs that solve specific problems. More complicated agents include human beings and organizations of human beings (such as firms). The paradigm gives researchers license to study isolated problems and find solutions that are both verifiable and useful, without agreeing on one single approach. An agent that solves a specific problem can use any approach that works — some agents are symbolic and logical, some are sub-symbolic neural networks and others may use new approaches. The paradigm also gives researchers a common language to communicate with other fields—such as decision theory and economics—that also use concepts of abstract agents. The intelligent agent paradigm became widely accepted during the 1990s.[]
Agent architectures and cognitive architectures
Researchers have designed systems to build intelligent systems out of interacting intelligent agents in a multi-agent system.[৯৮] A system with both symbolic and sub-symbolic components is a hybrid intelligent system, and the study of such systems is artificial intelligence systems integration. A hierarchical control system provides a bridge between sub-symbolic AI at its lowest, reactive levels and traditional symbolic AI at its highest levels, where relaxed time constraints permit planning and world modelling.[৯৯] Rodney Brooks' subsumption architecture was an early proposal for such a hierarchical system.[১০০]

In the course of 50 years of research, AI has developed a large number of tools to solve the most difficult problems in computer science. A few of the most general of these methods are discussed below.

Search and optimization

সম্পাদনা

Many problems in AI can be solved in theory by intelligently searching through many possible solutions:[১০১] Reasoning can be reduced to performing a search. For example, logical proof can be viewed as searching for a path that leads from premises to conclusions, where each step is the application of an inference rule.[১০২] Planning algorithms search through trees of goals and subgoals, attempting to find a path to a target goal, a process called means-ends analysis.[১০৩] Robotics algorithms for moving limbs and grasping objects use local searches in configuration space.[৬৯] Many learning algorithms use search algorithms based on optimization.

Simple exhaustive searches[১০৪] are rarely sufficient for most real world problems: the search space (the number of places to search) quickly grows to astronomical numbers. The result is a search that is too slow or never completes. The solution, for many problems, is to use "heuristics" or "rules of thumb" that eliminate choices that are unlikely to lead to the goal (called "pruning the search tree"). Heuristics supply the program with a "best guess" for the path on which the solution lies.[১০৫]

A very different kind of search came to prominence in the 1990s, based on the mathematical theory of optimization. For many problems, it is possible to begin the search with some form of a guess and then refine the guess incrementally until no more refinements can be made. These algorithms can be visualized as blind hill climbing: we begin the search at a random point on the landscape, and then, by jumps or steps, we keep moving our guess uphill, until we reach the top. Other optimization algorithms are simulated annealing, beam search and random optimization.[১০৬]

Evolutionary computation uses a form of optimization search. For example, they may begin with a population of organisms (the guesses) and then allow them to mutate and recombine, selecting only the fittest to survive each generation (refining the guesses). Forms of evolutionary computation include swarm intelligence algorithms (such as ant colony or particle swarm optimization)[১০৭] and evolutionary algorithms (such as genetic algorithms and genetic programming).[১০৮]

Logic[১০৯] is used for knowledge representation and problem solving, but it can be applied to other problems as well. For example, the satplan algorithm uses logic for planning[১১০] and inductive logic programming is a method for learning.[১১১]

Several different forms of logic are used in AI research. Propositional or sentential logic[১১২] is the logic of statements which can be true or false. First-order logic[১১৩] also allows the use of quantifiers and predicates, and can express facts about objects, their properties, and their relations with each other. Fuzzy logic,[১১৪] is a version of first-order logic which allows the truth of a statement to be represented as a value between 0 and 1, rather than simply True (1) or False (0). Fuzzy systems can be used for uncertain reasoning and have been widely used in modern industrial and consumer product control systems. Subjective logic[১১৫] models uncertainty in a different and more explicit manner than fuzzy-logic: a given binomial opinion satisfies belief + disbelief + uncertainty = 1 within a Beta distribution. By this method, ignorance can be distinguished from probabilistic statements that an agent makes with high confidence.

Default logics, non-monotonic logics and circumscription[৫১] are forms of logic designed to help with default reasoning and the qualification problem. Several extensions of logic have been designed to handle specific domains of knowledge, such as: description logics;[৪৫] situation calculus, event calculus and fluent calculus (for representing events and time);[৪৬] causal calculus;[৪৭] belief calculus; and modal logics.[৪৮]

Probabilistic methods for uncertain reasoning

সম্পাদনা

Many problems in AI (in reasoning, planning, learning, perception and robotics) require the agent to operate with incomplete or uncertain information. AI researchers have devised a number of powerful tools to solve these problems using methods from probability theory and economics.[১১৬]

Bayesian networks[১১৭] are a very general tool that can be used for a large number of problems: reasoning (using the Bayesian inference algorithm),[১১৮] learning (using the expectation-maximization algorithm),[১১৯] planning (using decision networks)[১২০] and perception (using dynamic Bayesian networks).[১২১] Probabilistic algorithms can also be used for filtering, prediction, smoothing and finding explanations for streams of data, helping perception systems to analyze processes that occur over time (e.g., hidden Markov models or Kalman filters).[১২১]

A key concept from the science of economics is "utility": a measure of how valuable something is to an intelligent agent. Precise mathematical tools have been developed that analyze how an agent can make choices and plan, using decision theory, decision analysis,[১২২] information value theory.[৫৭] These tools include models such as Markov decision processes,[১২৩] dynamic decision networks,[১২১] game theory and mechanism design.[১২৪]

Classifiers and statistical learning methods

সম্পাদনা

The simplest AI applications can be divided into two types: classifiers ("if shiny then diamond") and controllers ("if shiny then pick up"). Controllers do however also classify conditions before inferring actions, and therefore classification forms a central part of many AI systems. Classifiers are functions that use pattern matching to determine a closest match. They can be tuned according to examples, making them very attractive for use in AI. These examples are known as observations or patterns. In supervised learning, each pattern belongs to a certain predefined class. A class can be seen as a decision that has to be made. All the observations combined with their class labels are known as a data set. When a new observation is received, that observation is classified based on previous experience.[১২৫]

A classifier can be trained in various ways; there are many statistical and machine learning approaches. The most widely used classifiers are the neural network,[১২৬] kernel methods such as the support vector machine,[১২৭] k-nearest neighbor algorithm,[১২৮] Gaussian mixture model,[১২৯] naive Bayes classifier,[১৩০] and decision tree.[১৩১] The performance of these classifiers have been compared over a wide range of tasks. Classifier performance depends greatly on the characteristics of the data to be classified. There is no single classifier that works best on all given problems; this is also referred to as the "no free lunch" theorem. Determining a suitable classifier for a given problem is still more an art than science.[১৩২]

 
A neural network is an interconnected group of nodes, akin to the vast network of neurons in the human brain.

The study of artificial neural networks[১২৬] began in the decade before the field AI research was founded, in the work of Walter Pitts and Warren McCullough. Other important early researchers were Frank Rosenblatt, who invented the perceptron and Paul Werbos who developed the backpropagation algorithm.[১৩৩]

The main categories of networks are acyclic or feedforward neural networks (where the signal passes in only one direction) and recurrent neural networks (which allow feedback). Among the most popular feedforward networks are perceptrons, multi-layer perceptrons and radial basis networks.[১৩৪] Among recurrent networks, the most famous is the Hopfield net, a form of attractor network, which was first described by John Hopfield in 1982.[১৩৫] Neural networks can be applied to the problem of intelligent control (for robotics) or learning, using such techniques as Hebbian learning and competitive learning.[১৩৬]

Hierarchical temporal memory is an approach that models some of the structural and algorithmic properties of the neocortex.[১৩৭]

Control theory, the grandchild of cybernetics, has many important applications, especially in robotics.[১৩৮]

AI researchers have developed several specialized languages for AI research, including Lisp[১৩৯] and Prolog.[১৪০]

Evaluating progress

সম্পাদনা

In 1950, Alan Turing proposed a general procedure to test the intelligence of an agent now known as the Turing test. This procedure allows almost all the major problems of artificial intelligence to be tested. However, it is a very difficult challenge and at present all agents fail.[১৪১]

Artificial intelligence can also be evaluated on specific problems such as small problems in chemistry, hand-writing recognition and game-playing. Such tests have been termed subject matter expert Turing tests. Smaller problems provide more achievable goals and there are an ever-increasing number of positive results.[১৪২]

The broad classes of outcome for an AI test are: (1) Optimal: it is not possible to perform better. (2) Strong super-human: performs better than all humans. (3) Super-human: performs better than most humans. (4) Sub-human: performs worse than most humans.[তথ্যসূত্র প্রয়োজন] For example, performance at draughts is optimal,[১৪৩] performance at chess is super-human and nearing strong super-human (see Computer chess#Computers versus humans) and performance at many everyday tasks (such as recognizing a face or crossing a room without bumping into something) is sub-human.

A quite different approach measures machine intelligence through tests which are developed from mathematical definitions of intelligence. Examples of these kinds of tests start in the late nineties devising intelligence tests using notions from Kolmogorov complexity and data compression.[১৪৪] Two major advantages of mathematical definitions are their applicability to nonhuman intelligences and their absence of a requirement for human testers.

 
An automated online assistant providing customer service on a web page - one of many applications of artificial intelligence.

Artificial intelligence techniques are pervasive and are too numerous to list. Frequently, when a technique reaches mainstream use, it is no longer considered artificial intelligence; this phenomenon is described as the AI effect.[১৪৫]

Competitions and prizes

সম্পাদনা

There are a number of competitions and prizes to promote research in artificial intelligence. The main areas promoted are: general machine intelligence, conversational behavior, data-mining, driverless cars, robot soccer and games.

A platform (or "computing platform") is defined as "some sort of hardware architecture or software framework (including application frameworks), that allows software to run." As Rodney Brooks[১৪৬] pointed out many years ago, it is not just the artificial intelligence software that defines the AI features of the platform, but rather the actual platform itself that affects the AI that results, i.e., there needs to be work in AI problems on real-world platforms rather than in isolation.

A wide variety of platforms has allowed different aspects of AI to develop, ranging from expert systems, albeit PC-based but still an entire real-world system, to various robot platforms such as the widely available Roomba with open interface.[১৪৭]

Artificial intelligence, by claiming to be able to recreate the capabilities of the human mind, is both a challenge and an inspiration for philosophy. Are there limits to how intelligent machines can be? Is there an essential difference between human intelligence and artificial intelligence? Can a machine have a mind and consciousness? A few of the most influential answers to these questions are given below.[১৪৮]

Turing's "polite convention": We need not decide if a machine can "think"; we need only decide if a machine can act as intelligently as a human being. This approach to the philosophical problems associated with artificial intelligence forms the basis of the Turing test.[১৪১]

The Dartmouth proposal: "Every aspect of learning or any other feature of intelligence can be so precisely described that a machine can be made to simulate it." This conjecture was printed in the proposal for the Dartmouth Conference of 1956, and represents the position of most working AI researchers.[১৪৯]

Newell and Simon's physical symbol system hypothesis: "A physical symbol system has the necessary and sufficient means of general intelligent action." Newell and Simon argue that intelligences consist of formal operations on symbols.[১৫০] Hubert Dreyfus argued that, on the contrary, human expertise depends on unconscious instinct rather than conscious symbol manipulation and on having a "feel" for the situation rather than explicit symbolic knowledge. (See Dreyfus' critique of AI.)[১৫১][১৫২]

Gödel's incompleteness theorem: A formal system (such as a computer program) cannot prove all true statements.[১৫৩] Roger Penrose is among those who claim that Gödel's theorem limits what machines can do. (See The Emperor's New Mind.)[১৫৪]

Searle's strong AI hypothesis: "The appropriately programmed computer with the right inputs and outputs would thereby have a mind in exactly the same sense human beings have minds."[১৫৫] John Searle counters this assertion with his Chinese room argument, which asks us to look inside the computer and try to find where the "mind" might be.[১৫৬]

The artificial brain argument: The brain can be simulated. Hans Moravec, Ray Kurzweil and others have argued that it is technologically feasible to copy the brain directly into hardware and software, and that such a simulation will be essentially identical to the original.[৭৭]

Predictions and ethics

সম্পাদনা

Artificial Intelligence is a common topic in both science fiction and projections about the future of technology and society. The existence of an artificial intelligence that rivals human intelligence raises difficult ethical issues, and the potential power of the technology inspires both hopes and fears.

In fiction, Artificial Intelligence has appeared fulfilling many roles, including a servant (R2D2 in Star Wars), a law enforcer (K.I.T.T. "Knight Rider"), a comrade (Lt. Commander Data in Star Trek: The Next Generation), a conqueror/overlord (The Matrix), a dictator (With Folded Hands), a benevolent provider/de facto ruler (The Culture), an assassin (Terminator), a sentient race (Battlestar Galactica/Transformers), an extension to human abilities (Ghost in the Shell) and the savior of the human race (R. Daneel Olivaw in the Asimov's Robot Series).

Mary Shelley's Frankenstein considers a key issue in the ethics of artificial intelligence: if a machine can be created that has intelligence, could it also feel? If it can feel, does it have the same rights as a human? The idea also appears in modern science fiction, including the films I Robot, Blade Runner and A.I.: Artificial Intelligence, in which humanoid machines have the ability to feel human emotions. This issue, now known as "robot rights", is currently being considered by, for example, California's Institute for the Future, although many critics believe that the discussion is premature.[১৫৭] The subject is profoundly discussed in the 2010 documentary film Plug & Pray.[১৫৮]

Martin Ford, author of The Lights in the Tunnel: Automation, Accelerating Technology and the Economy of the Future,[১৫৯] and others argue that specialized artificial intelligence applications, robotics and other forms of automation will ultimately result in significant unemployment as machines begin to match and exceed the capability of workers to perform most routine and repetitive jobs. Ford predicts that many knowledge-based occupations—and in particular entry level jobs—will be increasingly susceptible to automation via expert systems, machine learning[১৬০] and other AI-enhanced applications. AI-based applications may also be used to amplify the capabilities of low-wage offshore workers, making it more feasible to outsource knowledge work.[১৬১]

Joseph Weizenbaum wrote that AI applications can not, by definition, successfully simulate genuine human empathy and that the use of AI technology in fields such as customer service or psychotherapy[১৬২] was deeply misguided. Weizenbaum was also bothered that AI researchers (and some philosophers) were willing to view the human mind as nothing more than a computer program (a position now known as computationalism). To Weizenbaum these points suggest that AI research devalues human life.[১৬৩]

Many futurists believe that artificial intelligence will ultimately transcend the limits of progress. Ray Kurzweil has used Moore's law (which describes the relentless exponential improvement in digital technology) to calculate that desktop computers will have the same processing power as human brains by the year 2029. He also predicts that by 2045 artificial intelligence will reach a point where it is able to improve itself at a rate that far exceeds anything conceivable in the past, a scenario that science fiction writer Vernor Vinge named the "singularity".[১৬৪]

Robot designer Hans Moravec, cyberneticist Kevin Warwick and inventor Ray Kurzweil have predicted that humans and machines will merge in the future into cyborgs that are more capable and powerful than either.[১৬৫] This idea, called transhumanism, which has roots in Aldous Huxley and Robert Ettinger, has been illustrated in fiction as well, for example in the manga Ghost in the Shell and the science-fiction series Dune.

Edward Fredkin argues that "artificial intelligence is the next stage in evolution," an idea first proposed by Samuel Butler's "Darwin among the Machines" (1863), and expanded upon by George Dyson in his book of the same name in 1998.[১৬৬]

Pamela McCorduck writes that all these scenarios are expressions of the ancient human desire to, as she calls it, "forge the gods".[]

  1. TOPIO:
  2. Definition of AI as the study of intelligent agents:
  3. The intelligent agent paradigm: The definition used in this article, in terms of goals, actions, perception and environment, is due to Russell & Norvig (2003). Other definitions also include knowledge and learning as additional criteria.
  4. Although there is some controversy on this point (see Crevier (1993, p. 50)), McCarthy states unequivocally "I came up with the term" in a c|net interview. (Skillings 2006)
  5. McCarthy's definition of AI:
  6. নীচে দেখুন Philosophy এর অধীনে Dartmouth proposal
  7. This is a central idea of Pamela McCorduck's Machines That Think. She writes: "I like to think of artificial intelligence as the scientific apotheosis of a venerable cultural tradition." (McCorduck 2004, পৃ. 34) "Artificial intelligence in one form or another is an idea that has pervaded Western intellectual history, a dream in urgent need of being realized." (McCorduck 2004, পৃ. xviii) "Our history is full of attempts—nutty, eerie, comical, earnest, legendary and real—to make artificial intelligences, to reproduce what is the essential us—bypassing the ordinary means. Back and forth between myth and reality, our imaginations supplying what our workshops couldn't, we have engaged for a long time in this odd form of self-reproduction." (McCorduck 2004, পৃ. 3) She traces the desire back to its Hellenistic roots and calls it the urge to "forge the Gods." (McCorduck 2004, পৃ. 340–400)
  8. আশাবাদ বলতে যা বোঝাচ্ছে তার মধ্যে যেমন আছে গোঁড়ার দিকের এই বিষয়ের গবেষকদের ভবিষ্যদ্বাণী (optimism in the history of AI দেখুন) তেমনি আধুনিক transhumanistsধারণা সমূহও আছে যেমন Ray Kurzweil.
  9. এখানে "বাধা বিপত্তি" বলতে যা বোঝাচ্ছে তার মধ্যে আছে the ALPAC report of 1966, the abandonment of perceptrons in 1970, the Lighthill Report of 1973 and the collapse of the lisp machine market in 1987.
  10. AI applications widely used behind the scenes:
  11. Pamela McCorduck (2004, pp. 424) writes of "the rough shattering of AI in subfields—vision, natural language, decision theory, genetic algorithms, robotics ... and these with own sub-subfield—that would hardly have anything to say to each other."
  12. This list of intelligent traits is based on the topics covered by the major AI textbooks, including:
  13. General intelligence (strong AI) is discussed in popular introductions to AI:
  14. AI in myth:
  15. Cult images as artificial intelligence: These were the first machines to be believed to have true intelligence and consciousness. Hermes Trismegistus expressed the common belief that with these statues, craftsman had reproduced "the true nature of the gods", their sensus and spiritus. McCorduck makes the connection between sacred automatons and Mosaic law (developed around the same time), which expressly forbids the worship of robots (McCorduck 2004, পৃ. 6–9)
  16. Humanoid automata:
    Yan Shi: Hero of Alexandria: Al-Jazari:
    • "A Thirteenth Century Programmable Robot"। Shef.ac.uk। সংগ্রহের তারিখ ২০০৯-০৪-২৫ 
    Wolfgang von Kempelen:
  17. Artificial beings:
    Jābir ibn Hayyān's Takwin: Judah Loew's Golem: Paracelsus' Homunculus:
  18. AI in early science fiction.
  19. এই যে অন্তর্দৃষ্টি, যে, ডিজিট্যাল কম্পিউটারের পক্ষে যে কোন আকারগত যুক্তিসজ্জার প্রক্রিয়াকে অনুকৃত করা সম্ভব, এটাই চার্চ-টুরিং থিসিস হিসাবে সুপরিচিত।
  20. Formal reasoning:
  21. AI's immediate precursors: See also History of artificial intelligence § Cybernetics and early neural networks. Among the researchers who laid the foundations of AI were Alan Turing, John Von Neumann, Norbert Wiener, Claude Shannon, Warren McCullough, Walter Pitts and Donald Hebb.
  22. Hegemony of the Dartmouth conference attendees:
  23. রাসেল এবং নরভিগ লিখেছেন "কোন কম্পিউটার যখন চালাক-চতুর কিছু করে তখন তা তো অদ্ভুত মানতেই হবে।" Russell ও Norvig 2003, পৃ. 18
  24. "Golden years" of AI (successful symbolic reasoning programs 1956-1973): The programs described are Daniel Bobrow's STUDENT, Newell and Simon's Logic Theorist and Terry Winograd's SHRDLU.
  25. DARPA pours money into undirected pure research into AI during the 1960s:
  26. AI in England:
  27. Optimism of early AI:
  28. See History of artificial intelligence § The problems
  29. First AI Winter, Mansfield Amendment, Lighthill report
  30. Expert systems:
  31. Boom of the 1980s: rise of expert systems, Fifth Generation Project, Alvey, MCC, SCI:
  32. Second AI winter:
  33. Formal methods are now preferred ("Victory of the neats"):
  34. McCorduck 2004, পৃ. 480–483
  35. DARPA Grand Challenge – home page
  36. "Welcome"। Archive.darpa.mil। সংগ্রহের তারিখ ২০১১-১০-৩১ 
  37. Markoff, John (১৬ ফেব্রুয়ারি ২০১১)। "On 'Jeopardy!' Watson Win Is All but Trivial"The New York Times 
  38. Kinect's AI breakthrough explained
  39. Problem solving, puzzle solving, game playing and deduction:
  40. Uncertain reasoning:
  41. Intractability and efficiency and the combinatorial explosion:
  42. Psychological evidence of sub-symbolic reasoning:
  43. Knowledge representation:
  44. Knowledge engineering:
  45. Representing categories and relations: Semantic networks, description logics, inheritance (including frames and scripts):
  46. Representing events and time:Situation calculus, event calculus, fluent calculus (including solving the frame problem):
  47. Causal calculus:
  48. Representing knowledge about knowledge: Belief calculus, modal logics:
  49. Ontology:
  50. Qualification problem: While McCarthy was primarily concerned with issues in the logical representation of actions, Russell ও Norvig 2003 apply the term to the more general issue of default reasoning in the vast network of assumptions underlying all our commonsense knowledge.
  51. Default reasoning and default logic, non-monotonic logics, circumscription, closed world assumption, abduction (Poole et al. places abduction under "default reasoning". Luger et al. places this under "uncertain reasoning"):
  52. Breadth of commonsense knowledge:
  53. Dreyfus ও Dreyfus 1986
  54. Gladwell 2005
  55. Expert knowledge as embodied intuition:
  56. Planning:
  57. Information value theory:
  58. Classical planning:
  59. Planning and acting in non-deterministic domains: conditional planning, execution monitoring, replanning and continuous planning:
  60. Multi-agent planning and emergent behavior:
  61. Learning:
  62. Alan Turing discussed the centrality of learning as early as 1950, in his classic paper Computing Machinery and Intelligence. (Turing 1950)
  63. (pdf scanned copy of the original) (version published in 1957, An Inductive Inference Machine," IRE Convention Record, Section on Information Theory, Part 2, pp. 56-62)
  64. Reinforcement learning:
  65. Computational learning theory:
  66. Natural language processing:
  67. Applications of natural language processing, including information retrieval (i.e. text mining) and machine translation:
  68. Robotics:
  69. Moving and configuration space:
  70. Robotic mapping (localization, etc):
  71. Machine perception:
  72. Computer vision:
  73. Speech recognition:
  74. Object recognition:
  75. Emotion and affective computing:
  76. Gerald Edelman, Igor Aleksander and others have both argued that artificial consciousness is required for strong AI. (Aleksander 1995; Edelman 2007)
  77. Artificial brain arguments: AI requires a simulation of the operation of the human brain A few of the people who make some form of the argument: The most extreme form of this argument (the brain replacement scenario) was put forward by Clark Glymour in the mid-70s and was touched on by Zenon Pylyshyn and John Searle in 1980.
  78. AI complete: Shapiro 1992, পৃ. 9
  79. Nils Nilsson writes: "Simply put, there is wide disagreement in the field about what AI is all about" (Nilsson 1983, পৃ. 10).
  80. Biological intelligence vs. intelligence in general:
    • Russell ও Norvig 2003, পৃ. 2–3, who make the analogy with aeronautical engineering.
    • McCorduck 2004, পৃ. 100–101, who writes that there are "two major branches of artificial intelligence: one aimed at producing intelligent behavior regardless of how it was accomplioshed, and the other aimed at modeling intelligent processes found in nature, particularly human ones."
    • Kolata 1982, a paper in Science, which describes McCathy's indifference to biological models. Kolata quotes McCarthy as writing: "This is AI, so we don't care if it's psychologically real"[১]. McCarthy recently reiterated his position at the AI@50 conference where he said "Artificial intelligence is not, by definition, simulation of human intelligence" (Maker 2006).
  81. Neats vs. scruffies:
  82. Symbolic vs. sub-symbolic AI:
  83. Haugeland 1985, পৃ. 255।
  84. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.38.8384&rep=rep1&type=pdf
  85. Pei Wang (২০০৮)। Artificial general intelligence, 2008: proceedings of the First AGI Conference। IOS Press। পৃষ্ঠা 63। আইএসবিএন 978-1-58603-833-5। সংগ্রহের তারিখ ৩১ অক্টোবর ২০১১ 
  86. Haugeland 1985, পৃ. 112–117
  87. Cognitive simulation, Newell and Simon, AI at CMU (then called Carnegie Tech):
  88. Soar (history):
  89. McCarthy and AI research at SAIL and SRI International:
  90. AI research at Edinburgh and in France, birth of Prolog:
  91. AI at MIT under Marvin Minsky in the 1960s :
  92. Cyc:
  93. Knowledge revolution:
  94. The most dramatic case of sub-symbolic AI being pushed into the background was the devastating critique of perceptrons by Marvin Minsky and Seymour Papert in 1969. See History of AI, AI winter, or Frank Rosenblatt.
  95. Embodied approaches to AI:
  96. Revival of connectionism:
  97. Computational intelligence
  98. Agent architectures, hybrid intelligent systems:
  99. Hierarchical control system:
  100. Subsumption architecture:
  101. Search algorithms:
  102. Forward chaining, backward chaining, Horn clauses, and logical deduction as search:
  103. State space search and planning:
  104. Uninformed searches (breadth first search, depth first search and general state space search):
  105. Heuristic or informed searches (e.g., greedy best first and A*):
  106. Optimization searches:
  107. Artificial life and society based learning:
  108. Genetic programming and genetic algorithms:
    • Luger ও Stubblefield 2004, পৃ. 509–530,
    • Nilsson 1998, chpt. 4.2.
    • Holland, John H. (১৯৭৫)। Adaptation in Natural and Artificial Systems। University of Michigan Press। আইএসবিএন 0262581116 
    • Koza, John R. (১৯৯২)। Genetic Programming। MIT Press। আইএসবিএন 0262111705  অজানা প্যারামিটার |subtitle= উপেক্ষা করা হয়েছে (সাহায্য)
    • Poli, R., Langdon, W. B., McPhee, N. F. (২০০৮)। A Field Guide to Genetic Programming। Lulu.com, freely available from http://www.gp-field-guide.org.uk/। আইএসবিএন 978-1-4092-0073-4  |publisher= এ বহিঃসংযোগ দেয়া (সাহায্য)
  109. Logic:
  110. Satplan:
  111. Explanation based learning, relevance based learning, inductive logic programming, case based reasoning:
  112. Propositional logic:
  113. First-order logic and features such as equality:
  114. Fuzzy logic:
  115. Subjective logic:
  116. Stochastic methods for uncertain reasoning:
  117. Bayesian networks:
  118. Bayesian inference algorithm:
  119. Bayesian learning and the expectation-maximization algorithm:
  120. Bayesian decision theory and Bayesian decision networks:
  121. Stochastic temporal models: Dynamic Bayesian networks: Hidden Markov model: Kalman filters:
  122. decision theory and decision analysis:
  123. Markov decision processes and dynamic decision networks:
  124. Game theory and mechanism design:
  125. Statistical learning methods and classifiers:
  126. Neural networks and connectionism:
  127. kernel methods such as the support vector machine, Kernel methods:
  128. K-nearest neighbor algorithm:
  129. Gaussian mixture model:
  130. Naive Bayes classifier:
  131. Decision tree:
  132. Classifier performance:
  133. Backpropagation:
  134. Feedforward neural networks, perceptrons and radial basis networks:
  135. Recurrent neural networks, Hopfield nets:
  136. Competitive learning, Hebbian coincidence learning, Hopfield networks and attractor networks:
  137. Hierarchical temporal memory:
  138. Control theory:
  139. Lisp:
  140. Prolog:
  141. The Turing test:
    Turing's original publication: Historical influence and philosophical implications:
  142. Subject matter expert Turing test:
  143. Game AI:
  144. Mathematical definitions of intelligence:
    • Jose Hernandez-Orallo (২০০০)। "Beyond the Turing Test"Journal of Logic, Language and Information9 (4): 447–466। ডিওআই:10.1023/A:1008367325700। সংগ্রহের তারিখ ২০০৯-০৭-২১ 
    • D L Dowe and A R Hajek (১৯৯৭)। "A computational extension to the Turing Test"Proceedings of the 4th Conference of the Australasian Cognitive Science jSociety। সংগ্রহের তারিখ ২০০৯-০৭-২১ 
    • J Hernandez-Orallo and D L Dowe (২০১০)। "Measuring Universal Intelligence: Towards an Anytime Intelligence Test"। Artificial Intelligence Journal174 (18): 1508–1539। ডিওআই:10.1016/j.artint.2010.09.006 
  145. "AI set to exceed human brain power" (web article)। CNN.com। ২০০৬-০৭-২৬। সংগ্রহের তারিখ ২০০৮-০২-২৬ 
  146. Brooks, R.A., "How to build complete creatures rather than isolated cognitive simulators," in K. VanLehn (ed.), Architectures for Intelligence, pp. 225–239, Lawrence Erlbaum Associates, Hillsdale, NJ, 1991.
  147. Hacking Roomba » Search Results » atmel
  148. Philosophy of AI. All of these positions in this section are mentioned in standard discussions of the subject, such as:
  149. Dartmouth proposal:
  150. The physical symbol systems hypothesis:
  151. Dreyfus criticized the necessary condition of the physical symbol system hypothesis, which he called the "psychological assumption": "The mind can be viewed as a device operating on bits of information according to formal rules". (Dreyfus 1992, পৃ. 156)
  152. Dreyfus' critique of artificial intelligence:
  153. This is a paraphrase of the relevant implication of Gödel's theorems.
  154. The Mathematical Objection: Making the Mathematical Objection: Refuting Mathematical Objection: Background:
    • Gödel 1931, Church 1936, Kleene 1935, Turing 1937
  155. This version is from Searle (1999), and is also quoted in Dennett 1991, পৃ. 435. Searle's original formulation was "The appropriately programmed computer really is a mind, in the sense that computers given the right programs can be literally said to understand and have other cognitive states." (Searle 1980, পৃ. 1). Strong AI is defined similarly by Russell & Norvig (2003, p. 947): "The assertion that machines could possibly act intelligently (or, perhaps better, act as if they were intelligent) is called the 'weak AI' hypothesis by philosophers, and the assertion that machines that do so are actually thinking (as opposed to simulating thinking) is called the 'strong AI' hypothesis."
  156. Searle's Chinese Room argument: Discussion:
  157. Robot rights:
    • Russell ও Norvig 2003, পৃ. 964
    • "Robots could demand legal rights"BBC News। ২১ ডিসেম্বর ২০০৬। সংগ্রহের তারিখ ৩ ফেব্রুয়ারি ২০১১ 
    Prematurity of: In fiction:
  158. Independent documentary Plug & Pray, featuring Joseph Weizenbaum and Raymond Kurzweil
  159. টেমপ্লেট:Ford 2009 The lights in the tunnel
  160. "Machine Learning: A Job Killer?"
  161. AI could decrease the demand for human labor:
  162. In the early 70s, Kenneth Colby presented a version of Weizenbaum's ELIZA known as DOCTOR which he promoted as a serious therapeutic tool. (Crevier 1993, পৃ. 132−144)
  163. Joseph Weizenbaum's critique of AI: Weizenbaum (the AI researcher who developed the first chatterbot program, ELIZA) argued in 1976 that the misuse of artificial intelligence has the potential to devalue human life.
  164. Technological singularity:
  165. Transhumanism:
  166. AI as evolution:
উদ্ধৃতি ত্রুটি: <references>-এ সংজ্ঞায়িত "Dartmouth conference" নামসহ <ref> ট্যাগ পূর্ববর্তী লেখায় ব্যবহৃত হয়নি।

টেমপ্লেট:John McCarthy navbox

টেমপ্লেট:Philosophy of science টেমপ্লেট:Philosophy of mind টেমপ্লেট:Robotics

টেমপ্লেট:Use dmy dates bn:কৃত্রিম বুদ্ধিমত্তা