Semester 1

Question 1

Accuracy?

Answer 1

(TP+TN)/(TP+TN+FP+FN)

Question 2

Bayes Theorem

Answer 2

P(A/B)=P(A)*P(B/A)/P(B)

Posterior=Prior*Likelihood/Evidence

Question 3

Bayesian vs frequentist statitics

Answer 3

Bayesian asumes that the data are true, and that the parameters don’t have a single value, but a range of plausible values

Question 4

Bernoulli distribution?

Answer 4

Single trial, only 2 possible outcomes.

Question 5

Convolutional Network?

Question 6

Decision Tree?

Question 7

DSFI?

Question 8

Error due to bias?

Answer 8

Model doesn’t learn the patterns of the training data. It oversimplifies the patterns or learns the wrong patterns

Question 9

Error due to variance?

Answer 9

Model pays too much attention to patterns in the training data and “memorizes” the patterns. The model isn’t flexible enough to generalize into new data

Question 10

Gamma distribution?

Answer 10

Continuous probability distribution that models right-skewed data

Question 11

Gradient Boosting Machine

Question 12

How to measure dependence between variables if the variables are continuous? Discrete?

Answer 12

1. Pearson.
2. Spearman.

Question 13

KNN?

Answer 13

K Nearest Neighbours. Classification or Regression. Finds the K nearest points to a given new data point and uses the majority or average vote to predict the target variable.

Question 14

Likelihood estimation method?

Answer 14

What is the likelihood of the variable having a certain distribution, given the observed data?

Question 15

Precision?

Answer 15

TP/(TP+FP)

Question 16

Medida de qué tan hacia los extremos está una cierta distribución?

Answer 16

Kurtosis

Question 17

Recall?

Answer 17

TP/(TP+FN)

Question 18

Skewness?

Answer 18

Si la distribución está cargada hacia la izquierda o la derecha.

Question 19

Tipos de Naive Bayes?

Answer 19

Gaussian: continuous variables
Multinomial: discrete variables
Binary: binary variables

Question 20

Valores de distribución Normal, para +-1d.e? +-2d.e? +-3d.e?

Answer 20

34.1% (68.2%)
13.6% (95.4%)
2.1% (99.7%)

Question 21

Poisson distribution?

Answer 21

Models the probability of an event happening k times whithin a given interval of time or space.

Question 22

False Positive Rate?

Answer 22

1-Specificity = FP/(FP+TN)

Question 23

True Positive Rate?

Answer 23

TP/(TP+FN)

Question 24

Specificity?

Answer 24

TN/(TN+FP)

Question 25

Formula de Probabilidades? Odds? Como pasar de Odds a Probabilidades?

Answer 25

a. F/(F+NF).
b. F/NF.
c. b/b+1

Question 26

Neural Network?

Answer 26

Input layers + hidden layers + output layers.

Each neuron has a weight for each past neuron + one bias + function (usually sigmoid).

It calculates a Cost function and then looks for the negative gradient, which tells us which weights and biases we need to change in order to lower the cost, this is called Backpropagation.

Question 27

Stochastic gradient descent?

Answer 27

Stochastic means doing the gradient descent process but not with all samples at the same time, instead you compute it for one batch, take a step and then compute it for the next batch, and take another step, etc.