Math & Statistics - Machine Learning

Question 1

What are the components of machine learning?

Answer 1

Task T: example - playing checkers
Performance measure P: example - percentage of games won agains opponents
Training experience E: playing practice games against itself.

Question 2

What is the inductive learning hypothesis?

Answer 2

Any hypothesis found to approximate the target function well over a sufficiently large set of training examples will also approximate the target function well over other unobserved examples.

Question 3

How to acquire a Concept learning?

Answer 3

Acquiring the definition of a general category by samples of positive and negative training examples of the category.

Question 4

What is Concept Learning?

Answer 4

Concept learning can be viewed as the task of searching through a large space of hypotheses implicitly defined by the hypothesis representation. The goal of this search is to find the hypothesis that best fits the training examples.

Question 5

Algorithms to find Hypothesis that fit the Concept=1

Answer 5

1. FIND-S

2. Candidate Elimination

Question 6

What is the difference between parametric and non-parametric approaches fo find f( )?

Answer 6

non-parametric approaches
completely avoid this danger, since essentially no assumption about the
form of f is made.

Question 7

What is the difference between supervised and unsupervised statistical learning problems?

Answer 7

x

Question 8

What is another name used for qualitative variables?

Answer 8

Categorical

Question 9

How problems of categorical variables are classified?

Answer 9

We tend to refer to problems

with a quantitative response as regression problems

Question 10

How problems of quantitative variables are classified?

Answer 10

qualitative response are often referred to as classification problems

Question 11

How to measure the performance of a statistical learning method?

Answer 11

In order to evaluate the performance of a statistical learning method on a given data set, we need some way to measure how well its predictions actually match the observed data. That is, we need to quantify the extent
to which the predicted response value for a given observation is close to the true response value for that observation. In the regression setting, the most commonly-used measure is the mean squared error (MSE).

Question 12

What is supervised statistical learning?

Answer 12

Building a statistical model to predict, or estimating, an output based on one or more inputs.

Question 13

What is unsupervised statistical learning?

Answer 13

Learn relationships and structure from data, not trying to predict an output number.

Question 14

What is the difference between classification problems and quantitative problems?

Answer 14

Quantitative problems output a number, while classification problems output a classification.

Question 15

Data descoberta/invenção métodos de statistical learning?

Answer 15

Século 19 - Regressão Linear - Legendre e Gauss - Method of Least Squares
1940 - Logistic Regression - vários autores
1970 - Generalized Linear Models - Nelder e Wedderburn
1985 - Classification and Regression Trees - Breiman, Friedman, Olshen and Stone
1986 - Generalized Additive Models (non-linear models) - Hastie e Tibshirani

Question 16

Notation used by Introduction to Statistical Learning

Answer 16

n - represent the number of distinct points, or observations, in our sample.
p - denote the number of variables that are available for use in making predictions.
Color fonte (red) - variables
i - index the samples of observations. (from 1 to n)
j - index the variables (from 1 to p)
X - denote a n x p matrix whose (i, j)th element is xij. Like a spreadsheet.
Xt - transpose of matrix.
Yi - denote the ith observation of the variable on which we wish to make predictions.

Question 17

E

Answer 17

Random error term

Question 18

The accuracy of Y (chapeu) as prediction for Y depends on which two quantities?

Answer 18

1. Reducible error and

2. Irreducible error

Question 19

Why is the irreducible error larger than zero?

Answer 19

The quantity may contain unmeasured variables that are useful in predicting Y : since we don’t measure them, f cannot use them for its prediction.

Question 20

Reasons to estimate f

Answer 20

1. Prediction

2. Inference

Question 21

Modeling for inference

Answer 21

Descobrir o relacionamento entre as diferentes variáveis e o resultado (Y).

Question 22

Modeling for prediction

Answer 22

Gera uma previsão (Y) baseado em input X.

Question 23

What are common approaches to determine f?

Answer 23

Linear Regression

Logistic Regression