On the Importance of Teaching Professional Ethics to Computer Science and Engineering Students - PowerPoint PPT Presentation

Loading...

PPT – On the Importance of Teaching Professional Ethics to Computer Science and Engineering Students PowerPoint presentation | free to download - id: 68b5ed-NDdhZ



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

On the Importance of Teaching Professional Ethics to Computer Science and Engineering Students

Description:

Research Thinking and Writing Toolbox DVA403 Artifactual and Natural Intelligence Symbolic, Sub-symbolic and Agent-based Gordana Dodig Crnkovic – PowerPoint PPT presentation

Number of Views:17
Avg rating:3.0/5.0
Slides: 26
Provided by: GordanaDo81
Learn more at: http://www.idt.mdh.se
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: On the Importance of Teaching Professional Ethics to Computer Science and Engineering Students


1
Research Thinking and Writing Toolbox DVA403
Artifactual and Natural Intelligence
Symbolic, Sub-symbolic and Agent-based
Gordana Dodig Crnkovic School of Innovation,
Design and Engineering, Mälardalen University,
Sweden
2
Thinking and Intelligence
  • In our course, Research Thinking and Writing
    Toolbox, research thinking, along with writing,
    is a central topic.
  • Research thinking as thinking in general are
    based on a set of abilities that we call
    intelligence, so let us start from learning some
    basics about how we today understand
    intelligence, the ways we think, acquire
    knowledge and produce knowledge.

3
What is Intelligence?
4
Intelligence
  • This general ability is defined as a combination
    of a several specific abilities, which include
  • Adaptability to changes in the environment
  • Learning capacity for knowledge/skill
    acquisition
  • Capacity for reasoning and abstract thought
  • Ability to comprehend relationships/patterns/rule
    s
  • Ability to evaluate and judge
  • Capacity for original and productive thought
  • .

5
Intelligence
Howard Gardner's theory of multiple intelligences
identifies at least eight different components
logical, linguistic, spatial, musical,
kinesthetic, interpersonal, intrapersonal and
naturalist intelligence. IQ tests address only
linguistic and logical plus some aspects of
spatial intelligence, while other forms of
intelligence have been entirely ignored.
6
Artifactual/Artificial Intelligence
  • In an artifact, artifactual/artificial
    intelligence is such a behavior (function) which
    in humans would require (biological)
    intelligence.
  • The central functions include reasoning,
    knowledge, planning, learning, communication,
    perception and locomotion (movement).

7
Artifactual/Artificial Intelligence
  • Artificial Intelligence (AI) is the branch of
    computer science that aims to create the
    intelligence of artifacts/ machines. John
    McCarthy coined the term AI in 1956.
  • Weak AI refers to the use of software to
    specific problem solving, (e.g. expert systems).
  • General intelligence (or Strong AI") is still a
    long-term goal of AI research (human-like
    intelligence).

8

Artifactual/Artificial Intelligence
  • In the beginning researchers started from human
    intelligence and tried to implement corresponding
    functions into machines (artifacts).
  • The problem was that no adequate understanding of
    human intelligence was available at that time.

9

Symbolic Intelligence Deduction, Reasoning and
Problem Solving
Human ability to think was the first thing AI
researchers tried to simulate. Early AI developed
algorithms that mimicked the step-by-step
reasoning that humans use to make logical
deductions. However, soon it was evident that
deduction is not enough.
10

Symbolic Intelligence Deduction, Reasoning and
Problem Solving
A very central itelligent ability that human
possess is our skill to handle uncertainty and
incomplete (often even contradictory)
information. Exact reasoning leads to the
explosion of possible scenarios which must be
analysed known as combinatorial explosion.
11

Symbolic Intelligence Deduction, Reasoning and
Problem Solving
A big advantage of machines their ability to
perform exact and lengthy calculations is at the
same time their problem in real life we do not
think perfectly exactly, but good enough.
Humans are taking into account relevant things,
and neglecting irrelevant. How can machine know
what is relevant?
12

Symbolic Intelligence Deduction, Reasoning and
Problem Solving
Symbolic information processing reasoning, on
the level of language (natural or formal), that
which we are aware of. Sub-symbolic information
processing on the level of electrical/chemical
signals, that which goes on in our brains and
nervous system without our thinking of it
seeing, motion, feelings, etc.
13

Symbolic Intelligence Deduction, Reasoning and
Problem Solving
Humans usually solve problems using fast,
intuitive judgments (feeling) on a level of
sub-symbolic information processing rather than
step-by-step deduction from perfectly exact data.
14

Symbolic Intelligence Deduction, Reasoning and
Problem Solving
Imitating sub-symbolic problem solving embodied
agent approaches emphasize the importance of
sensorimotor skills to higher reasoning neural
networks (connectionist) research simulates the
structures inside human and animal brains that
give rise to this sub-symbolic skill.
15
The Symbol Grounding Problem
  • GOFAI Good Old-Fashioned Artificial Intelligence
    is an ironic description of the oldest original
    approach to AI, based on logic and problem
    solving in specific problem domains, for example
    chess playing.
  • The term "GOFAI" was coined by John Haugeland in
    his 1986 book Artificial Intelligence The Very
    Idea, which explored the philosophical
    implications of artificial intelligence research.

16
The Symbol Grounding Problem
  • The GOFAI approach is based on the assumption
    that the most important aspects of intelligence
    can be achieved by the manipulation of symbols,
    known as the "physical symbol systems hypothesis"
    (Alan Newell and Herbert Simon in the middle
    1960s).

17
The Symbol Grounding Problem
  • GOFAI was the dominant paradigm of AI research
    from the middle 1950s until the late 1980s. The
    Symbol Grounding Problem is related to the
    problem of how words (symbols) get their
    meanings, and hence to the problem of what
    meaning itself really is.
  • If symbols (words) always are explained with
    other symbols we get infinite regress. Somewhere
    symbols must be grounded! In what way does
    that grounding happen?

18
Sub-symbolic AI
  • Opponents of the symbolic AI include roboticists
    such as Rodney Brooks, who construct autonomous
    robots without symbolic representation and
    computational intelligence researchers, who apply
    techniques such as neural networks to solve
    problems in machine learning and control
    engineering.
  • http//www.youtube.com/watch?vVyzVtTiax80NR1
    Self-Replicating Repairing Robots
  • http//www.youtube.com/watch?vTq8Yw19bn7Q Robots
    inspired by animals
  • http//www.youtube.com/watch?vO5DIyUWR-YYfeatu
    rerelated Rodney Brooks

19
Connectionist AI
  • Connectionist AI systems are large networks of
    extremely simple numerical processors, massively
    interconnected and running in parallel. The level
    of analysis at which uniform formal principles of
    cognition can be found is the subsymbolic level,
    intermediate between the neural and symbolic
    levels. Symbolic level structures provide only
    approximate accounts of cognition. Paul Smolensky
    http//web.jhu.edu/cogsci/people/faculty/Smolensky
    /

20
Connectionist AI
  • The Blue Brain Project simulation by
    reverse-engineering the mammalian brain.
    http//bluebrain.epfl.ch/

21
Connectionist AI
A model of brains  neocortical column, with a
generic facility that could allow modeling, and
simulation of any brain region for which the data
are provided.
http//www.hiddengarments.cn/?tagswitzerland
22
Integrating the Approaches Intelligent Agent
Paradigm
  • Nowadays, the term agent is used to indicate
    entities ranging all the way from simple pieces
    of software to "conscious" entities with learning
    capabilities.
  • For example, there are "helper" agents for web
    retrieval, robotic agents to explore inhospitable
    environments, agents in an economy, and so forth.

23
Integrating the Approaches Intelligent Agent
Modelling
  • An "agent" must be identifiable, that is,
    distinguishable from its environment by some kind
    of spatial, temporal, or functional attribute.
  • Moreover, agents must have some autonomy of
    action and they must be able to engage in tasks
    in an environment without direct external
    control.

24
Agent Based Modelling Approach
  • Agent-Based Modeling (ABM), a relatively new
    computational modeling paradigm, is the modeling
    of phenomena as dynamical systems of interacting
    agents. Another name for ABM is individual-based
    modeling.
  • This strongly resembles Marvin Minskys ideas of
    The Society of Mind and Douglas Hofstadters
    ideas about reductionism vs holism from his book
    Gödel, Escher, Bach An Eternal Golden Braid.

25
References
  • Basic material
  • http//en.wikipedia.org/wiki/Artificial_intelligen
    ce
  • http//paul-baxter.blogspot.com/2007/01/lessons-fo
    r-symbolic-and-sub-symbolic.html
  • http//en.wikipedia.org/wiki/Society_of_Mind
  • http//www.scholarpedia.org/article/Agent_based_mo
    deling
  • http//cogprints.org/3106/1/sgproblem1.html
    Harnad, S. (1990) The Symbol Grounding Problem.
    Physica D 42 335-346.
  • http//www.typos.de/pdf/2007_AI_without_representa
    tion_MM.pdf Vincent C. Müller, Is there a future
    for AI without representation?Minds and Machines,
    17 (1), 101-15.

25
About PowerShow.com