- What is Machine Learning
- Machine Learning Life Cycle
- Training-Validation-Error
- Model Training
- Machine Learning is the science (and art) of programming computers so they can learn from data.
- Here is a slightly more general definition:
Machine Learning is the field of study that gives computers the ability to learn without being explicitly programmed.
— Arthur Samuel, 1959
- And a more engineering-oriented one:
A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E.
— Tom Mitchell, 1997
Here the `rules` is difined as `models`
To learn the parameters, we follow this paradigm
- Collect lots of data pairs (Input Vector, Output Vector) =
$\large{\color{Purple} (x, y)}$ - Guess for the form of the hypothesis function
$\large{\color{Purple} h(x; w)}$ - Example :
$\large{\color{Purple}h(x; w)= w_0+w_1x_1 + w_2x_2}$
- Example :
- For an arbitrary guess for
$\large{\color{Purple}w}$ - We will get some
$\large{\color{Purple} \hat{y}= h(x; w)}$ ${\color{Blue}\textrm{which will not match the ground truth }} \large {\color{Purple} y}$
- We will get some
- Define a cost function
$\large{\color{Purple} J(y, \hat{y}(w))}$ depending on the difference. - Find optimal
$\large{\color{Purple} w}$ by minimizing$\large{\color{Purple} J(w)}$ by using some optimization procedure such as Gradient Descent.
- Machine Learning systems can be classified according to the amount and type of supervision they get during training.
- There are four major categories:
