Homeworks for BE5B33KUI - Cybernetics and Artificial Intelligence, given at CVUT@Prague.
An implementation on A* algorithm for state-space search.
Relevant files:
- agent.py
- heuristics.py
An implementation of alpha-beta pruning algorithm on the game Reversi (or Othello)
Relevant files:
- player.py
- myboard.py
Sequential decision processes under uncertainty. Implementations of Value and Policy iteration.
Relevant files:
- mdp_agent.py
Reinforcement learning in Gridworld. (Implemented Q-Learning)
Relevant files:
- rl_agent.py
Written report on letter recognition task. Source code for graphs only, as there is no need for a whole .tex environment here.
Relevant files:
- Assignment5OzanSan.pdf
- decider.m
- getConfusionMatrix.m
Letter recognition task. Implemented Naive Bayes and k-NN (specifically 1-NN) classifiers. Achieving around ~99% accuracy with 1-NN.
Relevant files:
- bayesLearn.m
- bayesClassify.m
- nnLearn.m
- nnClassify.m