Skip to content

Latest commit

 

History

History
61 lines (38 loc) · 1.64 KB

L01 Couse Introduction.md

File metadata and controls

61 lines (38 loc) · 1.64 KB

Learning goals of the course

  • Core methods of KBAI
  • Tasks addressed by KBAI
  • How KBAI agents use these methods to address these tasks
  • Relationship between AI and human cognition

What is AI and what are the limitations/challenges

Conundrums in AI

  1. AI agents have limited resources
  2. Local computation to solve problems with global constraints
  3. Logic is deductive but problems are not
  4. Dynamic world but limited knowledge
  5. Difficult to explain/ justify AI agent's decisions

Characteristics of AI problems

e.g. Think about how an AI predicts the traffic

  1. Knowledge arrives incrementally
  2. Recurring patterns
  3. Many levels of abstraction
  4. Computationally intractable
  5. World is dynamic but knowledge of the world is static
  6. World is open-ended but knowledge of the world is limited

Major components of knowledge-based AI

  • "Deliberation": Reasoning, learning and memory

Four schools of AI

quadrant

Examples:

quadrant_eg1

quadrant_eg2

  • ... and of course, boundaries are flexible...

Cognitive system

  • Knowledge-based AI: cognitive system(s) (can be more than one system) that interact with the world (input: perception / output: actions)

cognitive_system

Three-layered architecture:

- Metacognition: reasons about deliberation
- Deliberation: reasons about reaction
- Reaction: inteactions with the external world

An agent can be:

  • Reactive: output is the direct reaction to the input
  • Deliberative: involves deliberation: e.g. think and act to achieve a goal