AI Understanding — Cognitive Skills; AI Elements — Agent & Environment

AI Understanding — Cognitive Skills

What is Intelligence?

Intelligence can be loosely defined as the capability to obtain knowledge and skills and to apply those in various situations without supervision.

AI Programming Cognitive Skills (Components of AI)

5.1 Learning

• The trial-and-error method is fundamental to AI learning.
• Includes memorizing individual items like different solutions to problems.

• Types of learning in AI:

Rote learning — storing results mechanically without understanding

• Learning by analogy — applying previously learned solutions to similar problems
• Inductive learning — generalizing from specific examples to general rules
• Deductive learning — applying general rules to specific cases

5.2 Reasoning

• Allows the platform to draw inferences that fit with the provided situation.
• Types of reasoning:
  • Deductive reasoning — from general rules to specific conclusions ("All men are mortal; Socrates is a man; therefore Socrates is mortal")
  • Inductive reasoning — from specific observations to general rules
  • Abductive reasoning — finding the most likely explanation
  • Probabilistic reasoning — reasoning under uncertainty using probability

5.3 Problem-Solving

• AI's problem-solving ability comprises analyzing data to find the solution by identifying relevant features.
• Techniques:
  • Search algorithms — BFS, DFS, A*

Constraint satisfaction — solve problems with constraints

• Planning and scheduling — sequence of actions
• Optimization algorithms — genetic algorithms, simulated annealing

5.4 Perception

• The element scans any given environment by using different sense-organs, either artificial or real.
• Examples:
  • Computer Vision — "seeing" through cameras
  • Speech recognition — "hearing" through microphones
  • Tactile sensing — robotic touch sensors
  • LIDAR — distance sensing for autonomous vehicles

5.5 Linguistic Intelligence (Language Understanding)

• Handles distinctive types of language over different forms of natural meaning.
• Enables AI to understand and generate human language.
• Foundation of NLP (Natural Language Processing).

AI Elements — Agent & Environment

What is an Agent?

An AI agent is a system that:

1. Perceives its environment through sensors

Acts upon that environment through actuators / effectors

The intelligence of agents is calculated by their ability to create goals and achieve them.

"Anything that can gather information about its environment and take action based on that information."

Types of Agents (by Type)

Types of Agents (by Complexity)

What is the Environment?

An environment in AI is the surrounding of the agent.

• The agent takes input from the environment through sensors.
• The agent delivers output to the environment through actuators.

Self-Driving Car — Agent & Environment Example

Environment: Roads, other vehicles, road signs, pedestrians, buildings, weather

SENSORS                    AI PROCESSING             ACTUATORS
---------                  -------------             ---------
Camera          -------->   Perception    -------->  Steering Wheel
GPS             -------->   Planning      -------->  Accelerator
Speedometer     -------->   Prediction    -------->  Brake
Accelerometer   -------->   Decision      -------->  Horn
IR Range Finder -------->               -------->  Turn Signals
LIDAR           -------->               -------->  Headlights

Properties of Environments