topics.txt

1. Introduction
I    Reasoning: Induction, deduction, abduction

2. Statistics
I    Descriptive and inferential statistics: Population, sample, statistic, parameters, estimator.
II    Population mean and standard deviation.
II    Sample estimate of mean and standard deviation.
II    Confidence interval for mean and proportion.
II    Sample size estimate for mean and proportion.   

3. Algorithms
I    Algorithmic complexity.
I    Understand an algorithm from description or Python code.
II    Time complexity function.
II    Best, average and worst case complexity.
II    Big-O notation.

4. Symbolic AI
I    Symbolic AI.
I    Forward and backward chaining.
I    Monte carlo search.
II    Boolean logic.
II    Rule-based systems and expert systems.

5. Natural language processing
I    Data representation in the computer.
I    Bag of words representation.
II    Term frequency-inverse document frequency (TF-IDF).

 
6. Machine learning
I    Types of machine learning problems.
I    Generalization: Training and test error.
II    Linear regression. Model, parameters, and cost function.

7. Image processing
II    Feature normalization and standardization.
II    K-means clustering. Model, cost function, parameters, and algorithm.

8. Optimization
II    Gradient descent algorithm.
I    Stochastic gradient descent.
II    Gradient of cost function.
II    Neural networks: Model (layers, activation functions), parameters, cost function.

9. Automatic differentiation
I    Automatic differentiation: Forward and reverse accumulation.
II    Computation graphs.
II    Automatic differentiation in Pytorch.
II    Implementation of neural networks in Pytorch.

10. Audio processing
I    Frequency spectrum and spectrogram.
II    Feature transformations and basis change.

 
11. Reinforcement learning
I    Reinforcement learning: Markov decision process (state, action, reward).
I    Epsilon-greedy action selection and optimistic initialization.
II    RL algorithms: Value iteration and Q-learning.
II    Optimal action and optimal policy.
II    Discount factor.
II    Value (of a state) and quality (of a state-action pair).

12. Causality
I    Statistical dependency.
II    Covariance and correlation coefficient.
I    Causality and confounders.
I    Randomized trials.

13. Algorithmic fairness
II    Fairness criteria: Demographic parity, equalized odds, equal opportunity.
I    Ethical challenges and dilemmas in AI