Foundation of Aritificial Intelligence

• Homework and final project of Foundation of Artificial Intelligence, Spring 2024, National Taiwan University.

HW1 and HW2 is from UC Berkeley Pacman Project

• UC Berkeley AI material
• about the search and multi-agent search

HW1 - Search:

• Students implement depth-first, breadth-first, uniform cost, and A* search algorithms. These algorithms are used to solve navigation and traveling salesman problems in the Pacman world.

HW2 - Multi-agent Search:

• Classic Pacman is modeled as both an adversarial and a stochastic search problem. Students implement multiagent minimax and expectimax algorithms, as well as designing evaluation functions.

HW3 and HW4 is for Machine Learning Part of the course

HW3 - Linear Model, Decision Tree, and Random Forest

• The goal of this programming part is to implement a supervised machine learning pipeline from scratch, which includes preprocessing, training, and evaluation of linear and nonlinear models on a given dataset.
• Apply this pipeline to a classification task and a regression task.
• Including preprocessing, models, training, and evaluation.
• Dataset: Iris dataset for classification and Boston Housing dataset for regression.

HW4 - Implement PCA, Autodecoder, DenosingAutoencoder

• With different observation of these 3 models
  • Reconstruct images.
  • Different Eigenvectors of PCA
  • Different Architecture of DenosingAutoencoder
  • Different Optimizers
  • Observation of loss history