Lab assignments in the course Artificial Intelligence at KTH Royal Institute of Technology (DD2380).
- Search (20p kattis)
- HMM(0-3)
- RL(35p kattis)
- Swish
- Planning
Using the minimax algorithm with alpha-beta pruning to find the best possible moves in a game of catching fish.
More information on how to set up the environment and run the code can be found in the search folder.
Using several algorithms to tackle hidden Markov problems.
HMM0 calculates the probability of the outcome of the next state given state probability, transformation probability matrix, and outcome probability matrix.
HHM1 calculates the probability of an observation sequence happening given the model by utilizing the forward algorithm(alpha-pass)
HHM2 finds the most likely sequence of the hidden states that results in a sequence of observation by utilizing the Viterbi algorithm
HMM3 estimates the model using the forward-backward algorithm
Using epsilon-greedy algorithm and Q-learning, we find an optimal policy.
More information on how to set up the environment and run the code can be found in the search folder.
Simple planning problem.
Online editor for this problem: Link
Simple logic problem.
Online editor for this problem: Link