Machine Learning

What is mearning learning

Tom M. Mitchell provided a widely quoted, more formal definition of the algorithms studied in the machine learning field: “A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P if its performance at tasks in T, as measured by P, improves with experience E.

E.g., Alaphgo

E is the experience of games alphgo played

P is the winning probability

T is the task of playing go games

Supervised machine learning

Learning from the training data set with the correct answer, then improve the performance. E.g., using twitter data to predict the sentiment.

Supervised learning problems are categorized into “regression” and “classification” problems.

Unsupervised machine learning

giving the machine data and expect the machine return the correct answer.

Unsupervised learning allows us to approach problems with little or no idea what our results should look like. We can derive structure from data where we don’t necessarily know the effect of the variables.

We can derive this structure by clustering the data based on relationships among the variables in the data.

Linear Regression

Givin a trainding data set to determin the hypothesis using like y = a + bx, it is one variable function. choose a and b to make the hyposhesis is close to the training data’s y.

Cost Function

Cost function is used to measure the average acuracy of the hyperthesis function.

Gradient Descent

An interate way to find the a and b. It needs to update a and b simultaneously. Learning rate must be set as an appropriate value.