Click here to Skip to main content
15,904,494 members
Home / Discussions / C#
   

C#

 
GeneralRe: C# CF image processing Pin
Christian Graus19-Apr-04 17:26
protectorChristian Graus19-Apr-04 17:26 
GeneralRe: C# CF image processing Pin
icem4n23-Apr-04 17:11
icem4n23-Apr-04 17:11 
GeneralAI programming using C#.Net Pin
sreejith ss nair19-Apr-04 4:09
sreejith ss nair19-Apr-04 4:09 
GeneralRe: AI programming using C#.Net Pin
Heath Stewart19-Apr-04 4:43
protectorHeath Stewart19-Apr-04 4:43 
GeneralRe: AI programming using C#.Net Pin
Tom Larsen19-Apr-04 6:36
Tom Larsen19-Apr-04 6:36 
GeneralRe: AI programming using C#.Net Pin
leppie19-Apr-04 9:43
leppie19-Apr-04 9:43 
GeneralRe: AI programming using C#.Net Pin
HAHAHA_NEXT19-Apr-04 11:42
HAHAHA_NEXT19-Apr-04 11:42 
GeneralRe: AI programming using C#.Net Pin
Mikko Puonti19-Apr-04 12:44
Mikko Puonti19-Apr-04 12:44 
There is some simple implementations for different kind AI algorithms here in codeproject:

  • Fuzzy Logic Dot Net[^]
  • Neural Dot Net Pt 1 Introduction[^]
  • Path Finder[^]
  • A Simple C# Genetic Algorithm
    [^]
  • Inference in Belief Networks
    [^]
  • MetaAgent, a Steering Behavior Template Library[^]


  • Basicly AI algorithms are divided to two categories: symbolic and connectionism.

    Connectionism (from Stanford Encyclopedia of Philosophy[^]):
    Connectionism is a movement in cognitive science which hopes to explain human intellectual abilities using artificial neural networks (also known as ‘neural networks’ or ‘neural nets’). Neural networks are simplified models of the brain composed of large numbers of units (the analogs of neurons) together with weights that measure the strength of connections between the units. These weights model the effects of the synapses that link one neuron to another. Experiments on models of this kind have demonstrated an ability to learn such skills as face recognition, reading, and the detection of simple grammatical structure.
    (Note: Neural networks are not only connectionism approach. Also there is very different kind neural networks.)

    And symbolic (from The University of Alberta's Dictionary of Cognitive Science
    [^] ):
    Symbolic architecture refers to the classical view of the architecture of the mind. In this approach the mind is viewed as a process in which symbols are manipulated. Symbols are moved between memory stores such as long term and short term memory and are acted upon by an explicit set of rules in a particular sequence. The symbolic architecture is the manner in which memory stores are related and the set of rules applied to the system.

    Symbolic approach have achieved huge success in very limited problem areas. It's solutions are normally very specific for one problem. And because problems are normally simplified "laboratory problems" - this approach haven't achieved great practical success.

    Connectionism approaches advantage is that it's basic problem solution mechanism (neural networks and other similar networks) is applicable to wide area of different kind problems and you don't need to know much from problem (basicly it is enough to know inputs and outputs - and guess right network architecture). However it uses lot of processing power and memory (it is simple simulation from brains). You also need to find suitable presentation for inputs and outputs, train network and select suitable network architecture (how many layers, how many neurons, etc.).

    When you know your problem very well: you should use symbolic approach (or better develope mathematical model). If your problem is complex, you don't know it specifics well enough or there is lot's of "noise data" - you can try connectionism approach. Lisp and Prolog are developed for symbolic AI approaches. Connectionism approach doesn't have specific language - you just need to have programming language (like C#) that is suitable for "brain simulation".

    IMHO: Connectionism approach is real way to do AI. Symbolic approach seem to be suitable for only very limited problem areas.

    Here is some good tutorials for neural networks: AI Depot tutorials[^]
    GeneralAI programming using C#.Net Pin
    Anonymous19-Apr-04 4:08
    Anonymous19-Apr-04 4:08 
    GeneralDirectory Picker Pin
    bertcox19-Apr-04 3:38
    bertcox19-Apr-04 3:38 
    GeneralRe: Directory Picker Pin
    Heath Stewart19-Apr-04 3:40
    protectorHeath Stewart19-Apr-04 3:40 
    GeneralRe: Directory Picker Pin
    bertcox19-Apr-04 4:03
    bertcox19-Apr-04 4:03 
    Generalc++ client for c# server Pin
    rana7419-Apr-04 2:04
    rana7419-Apr-04 2:04 
    GeneralRe: c++ client for c# server Pin
    Heath Stewart19-Apr-04 3:17
    protectorHeath Stewart19-Apr-04 3:17 
    GeneralRe: c++ client for c# server Pin
    rana7419-Apr-04 18:12
    rana7419-Apr-04 18:12 
    GeneralRe: c++ client for c# server Pin
    Heath Stewart20-Apr-04 3:17
    protectorHeath Stewart20-Apr-04 3:17 
    GeneralConvert from string to Color Pin
    Member 74394419-Apr-04 1:49
    Member 74394419-Apr-04 1:49 
    GeneralRe: Convert from string to Color Pin
    Heath Stewart19-Apr-04 3:22
    protectorHeath Stewart19-Apr-04 3:22 
    GeneralRe: Convert from string to Color Pin
    Member 74394419-Apr-04 3:40
    Member 74394419-Apr-04 3:40 
    GeneralRe: Convert from string to Color Pin
    Heath Stewart19-Apr-04 4:01
    protectorHeath Stewart19-Apr-04 4:01 
    GeneralConvert from string to Color Pin
    Member 74394419-Apr-04 1:44
    Member 74394419-Apr-04 1:44 
    GeneralRe: Convert from string to Color Pin
    Mike Ellison19-Apr-04 4:07
    Mike Ellison19-Apr-04 4:07 
    QuestionCan Fill? Pin
    wokankan19-Apr-04 1:33
    wokankan19-Apr-04 1:33 
    AnswerRe: Can Fill? Pin
    mhmoud rawas19-Apr-04 2:11
    mhmoud rawas19-Apr-04 2:11 
    GeneralRe: Can Fill? Pin
    wokankan19-Apr-04 2:45
    wokankan19-Apr-04 2:45 

    General General    News News    Suggestion Suggestion    Question Question    Bug Bug    Answer Answer    Joke Joke    Praise Praise    Rant Rant    Admin Admin   

    Use Ctrl+Left/Right to switch messages, Ctrl+Up/Down to switch threads, Ctrl+Shift+Left/Right to switch pages.