3 Main Types of ML
- Supervised
- Unsupervised
- Reinforcement Learning
Examples of Supervised ML Algorithms
- Linear Regression
- Logistic Regression
- KNN
- SVM
Examples of Unsupervised ML Algorithms
K-Means Clustering
Examples of Reinforcement Learning Algorithms
Q-Learning
SARSA
What is Reinforcement Learning?
An agent interacts with its environment by producing actions and discovers errors or rewards
What is the field of ML about?
Parsing data, learning from data to make informed decisions about data.
Classification vs Regression
- Classification is the task of predicting a discrete class label (e.g. email spam or non-spam)
- Regression is the task of predicting a continuous quantity (e.g. predicting stock market prices)
Selection bias
- Statistical error that occurs in sampling portion of experiment
- Means one group is sampled more than other groups
- Usually produces inaccurate results
Recall
TP / P
Precision
TP / TP + FP
Confusion Matrix
A confusion matrix or an error matrix is a table which is used for summarizing the performance of a classification algorithm.
Inductive vs Deductive Learning
Inductive - using observations to draw conclusions
(data -> model)
Deductive - using conclusions to form observations (model to predictions)
KNN vs K-Means Clustering
- kNN is supervised ML, classification/regression, k is the number of neighbours
- K-Means clustering is unsupervised ML where k is the number of clusters the algo is trying to identify
Type 1 Error
False Positive
Type 2 Error
False Negative
