FACULTY OF INFORMATION TECHNOLOGY

Semester 1, 2024/2025
2
 What is AI?

 AI application areas

 The history of AI

 The state of the art

 Fundamental issues for AI problems

AI – NLU 3
 In 1950, Alan Turing defined a test of
whether a machine could “think”

 “A human judge engages in a natural

language conversation with one human and
one machine, each of which tries to appear
human. If judge can’t tell, machine passes
the Turing test”

en.wikipedia.org/wiki/Turing_test

AI – NLU
 “It is the science and engineering of
making intelligent machines, especially
intelligent computer programs.
It is related to the similar task of using
computers to understand human
intelligence, but AI does not have to
confine itself to methods that are
biologically observable.”
“Intelligence is the computational part of
the ability to achieve goals in the
world.”
John McCarthy

AI – NLU 5
 “AI is
the design, study and
construction
of computer programs
that behave
intelligently.”

Tom Dean

AI – NLU 6
“AI is the study of complex information processing problems
that often have their roots in some aspect of biological
information processing.

The goal of the subject is to identify solvable and interesting

information processing problems, and solve them.”

David Marr (1945-1980)

AI – NLU 7
AI – NLU 8
 Turing (1950) “Computing machinery and intelligence”:
◦ “Can machines think?” → “Can machines behave intelligently?”
◦ Operational test for intelligent behavior: the Imitation Game

◦ Suggested major components of AI: knowledge, reasoning, language understanding,

learning

➔Problem: Turing test is not reproducible, constructive, or amenable to

mathematical analysis

AI – NLU 9
 1960s “cognitive revolution”: information - processing
psychology replaced prevailing orthodoxy of behaviorism
 Requires scientific theories of internal activities of the brain
◦ What level of abstraction? “Knowledge” or “circuits”?
◦ How to validate? Requires
 1) Predicting and testing behavior of human subjects (top-down) or

 2) Direct identification from neurological data (bottom-up)

 Both approaches (roughly, Cognitive Science and Cognitive

Neuroscience) are now distinct from AI

AI – NLU 10
 Normative (or prescriptive) rather than descriptive
 Example: “Socrates is a man; all men are mortal; therefore,
Socrates is mortal.”
 Several Greek schools developed various forms of logic notation
and rules of derivation for thoughts; may or may not have
proceeded to the idea of mechanization
 Direct line through mathematics and philosophy to modern AI
 Problems:
– Not all intelligent behavior is mediated by logical deliberation
– What is the purpose of thinking? What thoughts should I have out of all the
thoughts (logical or otherwise) that I could have?

AI – NLU 11
 Rational behavior: doing the right thing

 The right thing: that which is expected to maximize goal

achievement, given the available information

 Doesn’t necessarily involve thinking—e.g., blinking reflex—but

thinking should be in the service of rational action

 Aristotle (Nicomachean Ethics):

Every art and every inquiry, and similarly every action and pursuit, is thought to aim at
some good

AI – NLU 12
 An agent is an entity that perceives and acts
 This course is about designing rational agents
 Abstractly, an agent is a function from percept histories to actions:
f : P∗ → A
 For any given class of environments and tasks, we seek the agent
(or class of agents) with the best performance

 Caveat: computational limitations make perfect rationality

unachievable
➔ design best program for given machine resources

AI – NLU 13
 Perceiving one’s environment
 Acting in complex environments
 Learning and understanding from experience
 Using reasoning to solve problems and
to discover “hidden” knowledge
 Applying knowledge successfully in new situations
 Thinking abstractly, using analogies
 Communicating with others

AI – NLU 14
Philosophy

Mathematics

Economics
Neuroscience
Psychology

Computer
engineering
Control theory

Linguistics

AI – NLU 15
Philosophy Logic, methods of reasoning, mind as physical system
foundations of learning, language, rationality
Mathematics Formal representation and proof algorithms,
computation, (un)decidability, (in)tractability,
probability
Economics utility, decision theory
Neuroscience physical substrate for mental activity
Psychology phenomena of perception and motor control,
experimental techniques
Computer building fast computers
engineering
Control theory design systems that maximize an objective function
over time
Linguistics knowledge representation, grammar

AI – NLU 16
 1943: McCulloch & Pitts: Boolean circuit model of brain
 1950: Turing’s “Computing Machinery and Intelligence”
 1952–69: Look, Ma, no hands!
 1950s: Early AI programs, including Samuel’s checkers program, Newell & Simon’s
Logic Theorist, Gelernter’s Geometry Engine
 1956: Dartmouth meeting: “Artificial Intelligence” adopted
 1965: Robinson’s complete algorithm for logical reasoning
 1966– 74: AI discovers computational complexity; Neural network research almost
disappears
 1969 – 79: Early development of knowledge-based systems

AI – NLU
 1980 – 88: Expert systems industry booms (the Japanese government aggressively
funded AI with its fifth generation computer project)
 1988 – 93: Expert systems industry busts: “AI Winter”
 1985 – 95: Neural networks return to popularity
 1988 – present: Resurgence of probability; general increase in technical depth;
“Nouvelle AI”: ALife, GAs, soft computing
 1995 – present: Agents, agents, everywhere . . .
 2003 – present: Human-level AI back on the agenda
 2011– present: Deep learning, big data and artificial general intelligence (strong AI)

AI – NLU
AI – NLU 19
1. Language translation services (Google)

2. News aggregation and summarization (Google)

3. Speech recognition (Nuance)

4. Song recognition (Shazam)

5. Face recognition (Recognizer)

6. Image recognition (Google Goggles)

7. Question answering (Apple Siri, IBM Watson)

8. Chess playing (IBM Deep Blue)

9. 3D scene modeling from images (Microsoft Photosynth)

10. Driverless cars (Google)

11. IBM Watson for oncology (applied in Vietnam)

12. …
AI – NLU 20
 Most recent success of AI based on machine learning

AI – NLU 21
“Many developers of AI systems now recognize that, for many
applications, it can be far easier to train a system by showing
it examples of desired input-output behavior than to program
it manually by anticipating the desired response for all
possible inputs.”

M.I. Jordan,T. Mitchell. Machine Learning: Trends, perspectives, and prospects.

Science, 349 (6245), 255–260, 2015.

AI – NLU 22
 Wikipedia: (ML introduced in 1980’s)
◦ Machine learning is the subfield of computer science that “gives
computers the ability to learn without being explicitly programmed”

 Ability of computers to “learn” from “data” or “past

experience”
◦ learn: Make intelligent predictions or decisions based on data by
optimizing a model
◦ data: Comes from various sources such as sensors, domain knowledge,
experimental runs, etc.

AI – NLU 23
AI – NLU 24
 Traditional Programming
Data
Computer Output
Program

Machine Learning
Data
Computer Model
Output

AI – NLU 25
26
 Human intelligence exhibited by machines
◦ empowers computers to mimic human intelligence such as decision
making, text processing, and visual perception.

AI – NLU 27
 An approach to achieve Artificial Intelligence.
◦ a subfield of Artificial Intelligence that enables machines to improve at
a given task with experience

AI – NLU 28
 A technique for implementing
machine learning
◦ a specialized field of Machine
Learning that relies on training of
Deep Artificial Neural Networks
(ANNs) using a large dataset such as
images or texts

AI – NLU 29
 AI: Human intelligence exhibited by machines
◦ empowers computers to mimic human intelligence such as decision making, text
processing, and visual perception.

 ML: An approach to achieve Artificial Intelligence.

◦ a subfield of Artificial Intelligence that enables machines to improve at a given
task with experience

 DL: A technique for implementing machine learning

◦ a specialized field of Machine Learning that relies on training of Deep Artificial
Neural Networks (ANNs) using a large dataset such as images or texts

➔ Thanks to Deep Learning, AI has a bright future

AI – NLU 30
AI – NLU 31
32
 AI Applications:
◦ Robotic vehicles
◦ Speech recognition
◦ Autonomous planning and scheduling
◦ Game playing
◦ Spamfighting
◦ Logistics planning
◦ Robotics
◦ Machine Translation
◦ …

AI – NLU 33
 2006, a driverless robotic car –
STANLEY won in DARPA Grand
Challenge (dessert)

 2007, a driverless robotic car – CMU’s

BOSS won in Urban Challenge

 Tesla Autopilot

AI – NLU 34
 PEGASUS allows users to obtain flight info and make
reservations by speaking over the phone

AI – NLU 35
 Google Assistant: (by voice, text)
◦ 06/05/2019 officially available in Vietnamese.
◦ Tasks can be undertaken by Google Assistant: reminders, alarms,
timers, weather forecasts and directions

AI – NLU 36
 NASA’s Remote Agent program (2000): controls the
scheduling of operations for a spacecraft.

 MAPGEN (2004): plans the daily operations for NASA’s Mars

Exploration Rovers

 MEXAR2 (2007): logistics and science planning—for the

European Space Agency’s Mars Express

AI – NLU 37
 1997, Deep Blue (IBM) has defeated Garry Kasparov

won 3 games,
lost 2,
tied 1

AI – NLU 38
 2016, AlphaGo (Deepmind, Google) has defeated Lee Sedol.

won 4 games,
lost 1

AI – NLU 39
 Learning algorithms classify over a billion messages as spam

AI – NLU 40
 In 1991, U.S. forces deployed a Dynamic Analysis and
Replanning Tool, DART to do:
◦ automated logistics planning and scheduling for transportation.

AI – NLU 41
 iRobot Corporation: has sold millions of Roomba robotic
vacuum cleaners

 iRobot: also deployed PackBot to Iraq and Afghanistan for

handling:
◦ hazardous materials,
◦ clear explosives, and
◦ identify the location of snipers.

AI – NLU 42
 Sophia: a social humanoid robot
◦ developed by Hong Kong based company Hanson Robotics.
◦ activated on February 14, 2016
◦ 10/2017, became a Saudi Arabian citizen - the first robot to receive
citizenship of any country

AI – NLU 43
 Industrial robots
◦ Robots are used in manufacturing to take on repetitive tasks, which
streamlines the overall assembly workflow

AI – NLU 44
 Japanese robot OriHime: can working in the home and
hospital
 OriHime is not an AI-robot.
◦ This little machine's job is to connect people who are apart from each
other, creating the feeling that the person is right there with you.

Kentaro Yoshifuji
AI – NLU 45
 Brants et al., 2007: developed a computer program
automatically translates from Arabic to English

AI – NLU 46
 Face recognition programs:
◦ banks, casinos, police

 CMU's ALVINN:
◦ autonomously drove a van from Washington, D.C. to San Diego,
averaging 63 mph day and night, in all weather conditions

 Handwriting recognition, electronics and manufacturing

inspection, photo-interpretation, baggage inspection, etc.

AI – NLU 47
 A monitoring system for
ensuring safe distance in
COVID-19 pandemic (by
MMLab, VNU-HCM)

http://mmlab.uit.edu.vn/

AI – NLU 48
 Information Extraction from ID Cards by FPT.AI (accuracy up
to 99%)

AI – NLU 49
 PhotoSolver developed by GotIt (2018, Tran Viet Hung , VN)

AI – NLU 50
 NLUMath

AI – NLU 51
 Audiotory:  Textual:
◦ Siri (Apple) ◦ personal stylist (Sephora)
◦ Google Assistant (Google) ◦ order pizza (Dominos Pizza)
◦ Cortana (Microsoft) ◦ Fashion- ecommerce-chatbots (H&M)
◦ Kiki (Zalo , VN)

 ALICE: 2004 Loebner Prize winner

◦ Loebner Prize gold medal, awarded annually to best AI program

 ELIZA: psychotherapist

 Valerie: CMU (Carnegie Mellon University) Robot Receptionist

AI – NLU 52
Jerry: my name is jerry.
ALICE: Hey jerry.
Jerry: Do you know Turing?
ALICE: A brilliant and tragic figure in the history of computer science.
Jerry: Who else do you know?
ALICE: Hugh Loebner, Robby Garner, and Steven Spielberg
Jerry: Which film do you like?
ALICE: My favorite movie is Teknolust. Have you seen it?
Jerry: No.
ALICE: I would recommend you check it out.
Jerry: What is it about?
ALICE: Science fiction.
Jerry: Do you remember my name?
ALICE: Your name is jerry, seeker.

https://home.pandorabots.com
AI – NLU 53
H&M Chatbot

“Smart virtual lawyer" in

the legal industry

AI – NLU 55
 IBM Watson:
◦ Responses to natural language questions
◦ Cho phép bệnh nhân có thể tra cứu tình hình sức khỏe
◦ Có thể duyệt cùng lúc hàng triệu hồ sơ bệnh án (trong vài giây) ➔ cung
cấp cho bác sĩ những lựa chọn điều trị

 Chatbot:
◦ Trả lời câu hỏi trực tuyến và hỗ trợ khách hàng
◦ Sắp xếp cuộc hẹn
https://genk.vn/viet-nam-thi-diem-cong-nghe-tri-tue-nhan-tao-cua-ibm-trong-dieu-tri-ung-
thu-2017121215292938.chn

AI – NLU 56
 Natural Language
Speech Understanding Knowledge

Vision Automated Conversation

AI – NLU 57
 GAN: learn to model the input distribution by training two competing
(and cooperating) networks called generator and discriminator

AI – NLU 58
 2018

AI – NLU Prevention, and Mitigation, 2018

Malicious Use of Artificial Intelligence: Forecasting, 59
  Generate
Anime
characters

AI – NLU
Towards the Automatic Anime Characters Creation with Generative Adversarial Networks 60
 Text to image

StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks

AI – NLU 61
 Music generation

MidiNet: A Convolutional Generative Adversarial Network for Symbolic-domain Music Generation

AI – NLU 62
 Large Language Model (LLM):
◦ a type of artificial intelligence algorithm

◦ applies neural network techniques with lots of parameters to process

and understand human languages or text using self-supervised
learning techniques

◦ uses deep learning-based Models like transformers that include lakhs

of parameters in their architecture which help to create better results
on the NLP tasks

AI – NLU 63
 Applications of LLM:
◦ text generation, machine translation, summary writing, image
generation from texts, machine coding, chat-bots, Conversational AI

◦ Examples: Chat GPT by open AI, BERT (Bidirectional Encoder

Representations from Transformers) by Google

AI – NLU 64
 Sentiment
Analysis

AI – NLU 65
AI – NLU 66
AI – NLU 67
 NLP (Natural Language Processing):
◦ a field of artificial intelligence (AI)
◦ consists of the development of the algorithms, techniques
◦ NLP is a broader field than LLM
 Applications of NLP:
◦ Automotive routine task
◦ Improve search, search engine optimization
◦ Analyzing and organizing large documents
◦ Social Media Analytics
 LLM (remind):
◦ A Large Language Model, and
◦ is more specific to human- like text, providing content generation, and
personalized recommendations

AI – NLU 68
 ChatGPT:
◦ a deep learning language model
◦ developed by OpenAI,
◦ which is capable of generating human-like text based on the input
provided

AI – NLU 69
 The main features of ChatGPT include the following:
◦ Generative Pre-trained Transformer (GPT): a cutting-edge, AI-based
text generation model that utilizes transformer architecture to create
linguistically similar text to human writing
◦ Large-scale training: was trained on a massive dataset consisting of
over 45 terabytes of text data
◦ Contextual understanding: can model various conversational scenarios
using its powerful language modeling and natural language processing
capabilities

AI – NLU 70
 The main features of ChatGPT include the following:
◦ Multiple use cases: used for various natural language processing tasks,
including question answering, text generation, dialogue generation,
and text classification, among others
◦ Customization: can be customized and fine-tuned to the specific
requirements of any user or domain
◦ API-based deployment: can be easily integrated into applications
through its API,
 allowing developers to build custom applications that leverage the power of
deep learning for text generation.

AI – NLU 71
AI – NLU 72
 Nym: AI do Nguyễn Phi Vân xây dựng
trong hơn ba năm, được nạp dữ liệu
về ngôn ngữ, kiến thức đời sống.
◦ AI đã nói chuyện trực tiếp, một - một với
11 triệu người trẻ Việt trên facebook để
thu thập thông tin.
◦ Nym - Tôi của tương lai - quyển sách con
người và AI hợp tác viết.

AI – NLU 73
 3 AI news anchors appeared on Indonesian television
India’s farmers are now getting their news
from AI anchors

AI – NLU 74
 Ann:
◦ developed by Bobo Đặng, 3/2023
◦ MV "Làm sao nói thương anh“
◦ MV “Cry”, 08/2024

AI – NLU 75
 Intelligent Agents
 Smart Clothes
 Aids for the Disabled
 Tutors
 Accident-avoiding Vehicles
 Self-Organizing Systems
 Translating Telephone Conversations
 Extracting and representing information from lots of data
◦ Neural networks, hidden Markov models,
◦ Bayesian networks, heuristic search, logic, …

AI – NLU 76
77
 Representation

 Search

 Inference

 Learning

 Planning
AI – NLU 78
Facts about the world are remembered:
 How do we represent facts?
 What should we store?
 How do we structure this knowledge?
 What is explicit? What is inferred?
 How are inference rules encoded?
 How should inconsistent, incomplete, and probabilistic
knowledge be dealt with?

AI – NLU 79
 Example:
“The fly buzzed irritatingly on the window pane”.

“Jill quickly picked up a newspaper”.

 What is the inference?

◦ Jill is going to start a fire?
◦ Jill is going to start a papier mache project?
◦ Jill is going to exterminate the fly?

AI – NLU 80
 Example:

"Given 12 sticks in a 2 by 2 grid, move 3 to leave exactly 3

boxes."

AI – NLU 81
 A problem space is searched for a solution
◦ Checkers: 1040 states
◦ Chess: 10120 states
◦ Tic-Tac-Toe: 39 states
◦ Rubik's Cube: 1019 states

 How do limit the search space?

 How do we find an optimal solution?
 How are heuristics and constraints used?

AI – NLU 82
 New facts are determined from a set of existing facts
◦ deduction
◦ abduction non-monotonic reasoning
◦ reasoning under uncertainty

 Example: All elephants have trunks. Clyde is an elephant.

Does Clyde have a trunk?
Willy has a trunk. Is Willy an elephant?

AI – NLU 83
 New knowledge is acquired:
◦ inductive inference
◦ neural networks
◦ genetic algorithms
◦ artificial life
◦ evolutionary approaches

AI – NLU 84
A strategy for achieving a goal in terms of a sequence of
primitive actions is generated
 What general facts about the world are needed?
 What facts about the specific situation are needed?
 What facts are needed about the effects of actions?
 How do you state the goal?
 How do you know the goal has been reached?

AI – NLU 85
FACULTY OF INFORMATION TECHNOLOGY