Course Overview for Artificial Intelligence (CIS 479)
What is Artificial Intelligence?
- Machine mimic human behavior
- Self adjusting code/ learning
- Study idea that enable computers to be intelligent
Why Study Artificial Intelligence?
- Find out how humans think
- Extend programming
- Real World applications
Heuristic Programming
How can you tell if a machine is intelligent?
- Reasoning
- Self awareness
- Learning – The capacity to acquire and apply knowledge
Turing Test
- the questioner can not tell which is the computer and which the person, there is a 50/50 chance of picking incorrectly
Can/Should Computers replace human experts?
- At the current moment in time no!
- Expert systems should still be used by experts (people that understand the way things should work)
Good Problems For Artificial Intelligence
- Decomposable to easier problems
- Solution steps can be ignored or undone
- Predictable Problem Universe
- Good Solutions are obvious
- Internally consistent knowledge base (KB)
- Requires lots of knowledge or uses knowledge to constrain solutions
- Interactive
Goals for Artificial Intelligence (AI)
- Make computers more useful
- Understand Principles of Human Intelligence
Newell/Simon Physical Symbol Hypothesis
- Given enough symbols
- Given collection of operators on expression
- Necessary & Sufficient Conditions for intelligent behavior
Artificial Intelligence (AI) Technique
- Capture generalizations
- Understood by domain expert
- Easily modified to correct errors
- Used in lots of different situations
- Reduce its own size/bulk, this does not spend the entire time looking at the whole domain
Successful Artificial Intelligence (AI) Applications
- Usually operate in a restricted domain
- Tend to exhibit goal directed planning behavior
- Some reasoning or explanation mechanism
Taxonomy of Artificial Intelligence System (From Easiest to Hardest)
- Data acquisition
- Expert System (Automatic Programming, Problem Solving)
- Robotic Control
- Pattern Recognition
- Natural Language Processing
- Vision & Scene Analysis
Criteria for Success
- Turing test
- More relaxed criteria
- Is the task clearly defined?
- Is there an implemented procedure performing the task?
- Is there an identifiable set of regularities or constraints that the procedure uses to derive its power?