Basics Flashcards
Machine Learning
Machine learning is programming computers to optimize a performance
criterion using example data or past experience.
Expressed as Triplets
(T,P,E)
T - Task
P - Performance
E - Experience
ML Algorithms:
Data Model Parameter Loss Function Learning Algorithm
Components of ML:
Data
Model/Hypothesis - Obtained from data
Hypothesis Space - set of all possible hypothesis
Parameters - Constant values obtained from the model
Hyper Parameters - constant values assumed for the building of model
Optimization problem - reducing the error for values of the model
Decision Function - model used for future predictions
Supervised Learning
machine learning task of learning a function that maps an input to an output based on example input-output pairs
Examples
Linear Regression Logistic Regression KNN SVM Naive Bayes Decision Tree Neural Networks
Unsupervised Learning
machine learning algorithm used to draw inferences from datasets consisting of input data without labeled responses
Examples
Clustering
Parametric Model
- fixed set of parameters over size of data
- simple
- small dataset
- we have assumptions about the model
Examples
Linear Regression Logistic Regression Naive Bayes Perceptron Neural Networks
Non Parametric Model
- we have no or very poor assumptions about the model
Examples
SVM
KNN
Decision Tree
Linear Model
- linear relationship
Example:
Linear and Logistic Regression
Perceptron
Non Linear Model
Deep Learning algorithms
Multi layer Perceptron
