Artificial Intelligence Fundamentals

Overview

A collection of different labs completed for my Artificial Intelligence class (6.034). Specifically, I cover representations, methods, and architectures used to build applications, and to account for understanding and developing intelligence from a computational perspective - aka "artificial intelligence" (AI).

1. Search

Exploration of different graph search approaches and algorithms.

• Blind Search
  ◦  Depth-First Search
  ◦  Breadth-First Search

• Heuristic Search
  ◦  Hill Climbing
  ◦  Best-First Search
  ◦  Beam Search

• Optimal Search
  ◦  Branch & Bound
  ◦  Branch & Bound w/ A*

2. Games

Representing a game as a decision tree and performing search operations on the corresponding tree.

• Representation Methods
  ◦  Game Completion
  ◦  Move Generation
  ◦  Determining Heuristics

• Search Methods
  ◦  Cooperative Depth-First Search
  ◦  Ordinary Minimax Search
  ◦  Minimax Search w/ Alpha-beta Pruning
  ◦  Progressive Deepening

3. Constraint Satisfaction

Exploring different approaches to solve constraint satisfaction problems (CSPs).

• Forward Checking
  ◦  Neighbor Elimination
  ◦  DFS Solver

• Propagation
  ◦  Domain Reduction
  ◦  Generic Propagation
  ◦  Propagation through Singleton domains
  ◦  Propagation through Reduced domains

4. Rule-based Systems

Exploring the structure and application of production rule systems w/ data and rules.

• Goal Trees
  ◦  Family Relations
  ◦  Backward Chaining

5. Bayesian Inference

Understanding the different components of building and evaluating Bayesian networks.

• Probabilities
  ◦  Joint Prob.
  ◦  Marginal Prob.
  ◦  Conditional Prob.

• Independence
  ◦  Conditional Indep.
  ◦  Structurally Indep.

• Misc.
  ◦  # of Parameters
  ◦  Node Relationships

6. ID Trees & k-Nearest Neighbors

Understanding how to effectively classify and interpret information w/ ID trees and kNNs.

• ID Trees
  ◦  Point Classification
  ◦  Disorder Calculation
  ◦  Tree Construction
  ◦  Greedy Construction

• k-Nearest Neighbors
  ◦  Distance Metrics
  ◦  Neighbor Classification
  ◦  k-Selection and Validation

7. Neural Networks

Understanding how network models interpret input information and produce meaningful output.

• Forward Propagation
  ◦  Node Value Determination
  ◦  Binary Output Calculation

• Backward Propagation
  ◦  Gradient Descent
  ◦  Determine Dependencies
  ◦  Delta-B Calculation
  ◦  Weight Update

8. Support Vector Machines (SVMs)

Exploring another method of supervised point classification, with SVMs utilizing relevant training points.

• Boundaries
  ◦  Positiveness
  ◦  Classification
  ◦  Margin
  ◦  Gutter Constraint

• Supportiveness
  ◦  Validation of Alpha Sign and Equation

• Evaluation
  ◦  Accuracy
  ◦  Training/SVM Update

9. Adaptive Boosting

Implementing the adaboost algorithm to combine the strengths of different classifiers.

• Adaboost Methods
  ◦  Weight Initialization
  ◦  Weight Update
  ◦  Error Calculation
  ◦  Voting Power
  ◦  Determine Misclassifications
  ◦  Algorithm for N Rounds