slide1
Download
Skip this Video
Download Presentation
2013.

Loading in 2 Seconds...

play fullscreen
1 / 49

2013. - PowerPoint PPT Presentation


  • 143 Views
  • Uploaded on

Department of Computing and Automation, Faculty of Technical Sciences. Chair of Informatics. 2013. Artificial intelligence. Đorđe Obradović, Marko Joci ć. Agenda. What is AI? History of AI Problems of AI Applications or uses of AI. What is Artificial intelligence?.

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about ' 2013.' - arnon


An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
slide1

Department of Computing and Automation, Faculty of Technical Sciences

Chair of Informatics

2013.

Artificial intelligence

Đorđe Obradović, Marko Jocić

agenda
Agenda
  • What is AI?
  • History of AI
  • Problems of AI
  • Applications or uses of AI
what is artificial intelligence
What is Artificial intelligence?
  • The branch of computer science
  • Problem solving
  • The ability of a computer or other machine to perform those activities that are normally thought to require intelligence
  • The study and design of Intelligent Agent
what is artificial intelligence1
What is Artificial intelligence?

1956. John Mc Carty, MIT

The science and engineering of making intelligent machine

  • especially intelligent computer programs. It is related to the similar task of using computers to understand human intelligence.
what is artificial intelligence2
What is Artificial intelligence?

The branch of computer science concerned with making computers behave like humans.

or

A system that makes it possible for a machine to perform functions similar to human intelligence.

history of ai
History of AI
  • 1956. Term Artificial Intelligence introduced
  • Computer were solving word problem in algebra, proving logical theorems and speaking English.
  • In the early 80s AI research was received by the commercial success of Expert System.
  • In the 90s and early 21st century: AI achieved its greatest success.
history of ai 90s and beyond
History of AI (90s and beyond)
  • 1997. Deep Blue – chess-playing computer beats reigning world champion Garry Kasparov
    • Kasparov demanded a rematch because he thought the computer was cheating 
  • Deep Blue could evaluate 200 million positions per second
    • Today’s Intel Core 2 Duo CPU is capable of evaluating 8 million positions
history of ai 90s and beyond1
History of AI (90s and beyond)
  • 2005. Stanford robot – driving autonomously for 131 miles along an unrehearsed desert trail
                  • 2010. Google driverless car – driving in cities, and also racing on test courses. No accidents so far!
                  • 2011 – present - self-driven cars legalized in some states of USA
history of ai 90s and beyond2
History of AI (90s and beyond)
  • 2011. Watson – IBM’s question answering machine defeated two greatest champions of Jeopardy! quiz
  • capable of answering questions posed in natural language
  • 200 millions pages of data (4 TB of disk storage)
  • received prize of $1 million for victory 
problems of ai
Problems of AI

The problem of simulating (or creating) intelligence has been broken down into a number of specific sub-problems.

problems of ai1
Problems of AI
  • Deduction, reasoning, problem solving
    • Early AI researches developed Algorithms that imitated the step-by-step reasoning that human beings use when they solve puzzles, play board games or make logical deductions.
problems of ai2
Problems of AI
  • Knowledge Representation
    • Many of the problems machines are expected to solve will require extensive knowledge about the world.
problems of ai3
Problems of AI
  • Planning
    • Intelligent agents must be able to set goals and achieve them. They need a way to visualize the future and be able to make choices that maximaze the utility (or “value”) of the available choices.
problems of ai4
Problems of AI
  • Learning – Machine learning
    • The mathematical analysis of the machine learning algorithms and their performance is branch of theoretical computer science known ascomputational learning theory
problems of ai5
Problems of AI
  • Natural language processing
    • Natural language processing gives machines the ability to read and understand the languages that human beings speak.
problems of ai6
Problems of AI
  • Motion and manipulation
    • The field of robotics is closely related to AI. Intelligence is required for robots to be able to handle such tasks as object manipulation 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)
problems of ai7
Problems of AI
  • Perception
    • Machine perception is the ability to use input from sensors (such as cameras, microphones, sonar and others more exotic) to deduce aspects of the world.
slide18

How it works?

    • How human beings performs intelligence tasks?
    • Is it possible to use those abilities in computer programs?
    • How?
motivation
Motivation
  • Try to solve this equations FAST without using calculator:
motivation1
Motivation
  • If the car traveling at the speed of 61.34 km/h, for how many minutes will exceed 29.56 km?
motivation2
Motivation
  • Do we use some shortcuts?
  • For example..
motivation4
Motivation
  • Do we use some shortcuts?
  • For example..

If the car traveling at the speed of 61.34 km/h, for how many minutes will exceed 29.56 km?

61.34 km/h is approximately 60 km/h, which is about 1 km/min.

So it means that we will need about half an hour to arrive at the destination of 29.56 km.

motivation5
Motivation
  • If we can use such shortcuts, how we can use them in computer programs?
  • In the real applications length of 16cm is almost same as 16.01cm
  • If we insist on accuracy sometimes our program cam make wrong conclusion.
motivation6
Motivation

Example 1

(5,5)

p1

(1,4)

Intersection p1 and p2

(1,1)

(5,1)

p2

motiva tion
Motivation

Example 1

(5,5)

p1

PostGIS 8.4.4 functions:

contains, intersection

(1,4)

contains(

intersection(

st_geomfromewkt(\'LINESTRING(1 1, 5 5)\'),

st_geomfromewkt(\'LINESTRING(1 4, 5 1)\')

),

st_geomfromewkt(\'LINESTRING(1 4, 5 1)\')

)

(1,1)

(5,1)

p2

motivation7
Motivation

What is the date?

How many triangles are

shown on the picture.

Is this a car or house?

motivation8
Motivation

Arrange geometric shapes to occupy the

smallest area.

Allowed operations are move and rotate

motivation9
Motivation

Which artist is most often appears on the

list. How many times?

motivation10
Motivation

Find 5 differences

motivation11
Motivation

What is shown on the figure?

motivation12
Motivation

How many persons are shown on the picture?

slide34

Parameters

Model structure

Hidden concepts

What

Supervised

Unsupervised

Crisp or uncertain

Reinforcement learning

Planning

Reasoning

Deterministic

Prediction

Pattern recognition

Aim

Learning

Diagnostic

Prediction

Stochastic

Diagnostic

Online, offline

Types

Model

Active, passive

Problems

State space

Classification

regression...

Results

Domains

Finance

AI - SOFT COMPUTING

Tools

Robotics

Neural computing

Games, entertainment, sport

Fuzzy systems

Medicine

Probabilistic reasoning

Evolutionary computation

WEB

Swarm intelligence

Agricultureand industry

natural language processing
Natural language processing
  • Gives machines the ability to read and understand the languages human beings speak
  • Requires in-depth knowledge of the languages that are processed
  • The goal is often natural language understanding – enabling computers to derive meaning from human language input
speech recognition
Speech recognition
  • Technology that allows the computer to identify and understand words spoken by a person using a microphone or telephone
  • Many challenges:
    • different accent
    • slang
    • background noise
facial recognition
Facial recognition
  • Automatically identifying or verifying a person from a digital image or a video frame from a video source
  • Uses:
    • security systems
    • crime prevention
    • auto-focus (on modern

mobile devices and camers)

military
Military
  • Faster collection and examination of large amounts of information
  • Improving reaction and decision-making for more effective military actions
  • High-performance fighter aircraft
  • Training simulators
medicine
Medicine
  • Clinical decision support systems
  • Computer-aided medical diagnosis
  • Computer-aided interpretation of medical images (CT, MRI, ultrasound…)
and many others
And many others…
  • Finance – market predicting
  • Transportation – automatic gearboxes who utilize fuzzy logic
  • Music – can computer be a skillful musician?
  • Toys – artificial pets
  • Real estate – risk analysis
  • E-mail spam filtering
  • Games
demonstrations
Demonstrations
  • Demos of AI in games:
    • Are you faster then a computer?
    • Can you beat a simple chess AI?
  • Demos of AI in computer vision:
    • Real-time object tracking
    • Medicine – image segmentation
    • Optical character recognition (OCR)
demonstrations1
Demonstrations
  • Demo of AI in natural language processing:
    • Have a chat with a bot
  • Others:
    • Control a car with motion gestures
    • Control mouse pointer with random object
ad