Glossary ML

Question 1

K-nearest neighbour

Answer 1

Class predicition by majority vote

Question 2

Overfitting

Answer 2

Modeling error due to too complex model

Question 3

Weak learner

Answer 3

Classifier performing better than random guess

Question 4

Latent variable

Answer 4

Variable inferred from other variables

Question 5

Error backpropagation

Answer 5

Gradient following in a neural network/An approach to train artificial neural networks

Question 6

Reward

Answer 6

Feedback from the environment

Question 7

Variance of a classifier

Answer 7

Divergence of estimated prediction function

Question 8

Fitness

Answer 8

Value used for the selection process

Question 9

Support vector

Answer 9

Samples on the margin of the decision surface/Data point affecting the decision boundary

Question 10

Negative sample

Answer 10

Training data which is not part of the concept being learned

Question 11

k-means

Answer 11

Clustering method based on centroids

Question 12

Normal distribution

Answer 12

Continous PDF defined by mean vector and covariance matrix

Question 13

Posterior probability

Answer 13

Probability after observation/Conditional probability taking into account the evidence

Question 14

The Lasso

Answer 14

An approach to regression that results in variable selection

Question 15

Principal Component Analysis

Answer 15

An unsupervised method for dimensionality reduction

Question 16

Perceptron learning

Answer 16

Error driven method to compute weights in a single layer neural network/Method to find seperating hyperplanes

Question 17

Categorical distribution

Answer 17

Distribution of discrete stochastic variables

Question 18

Subspace

Answer 18

A space spanned by a set of linearly independent vectors

Question 19

Dropout

Answer 19

A method of regularization used in deep neural networks/An approach to train artificial neural networks

Question 20

EM-algorithm

Answer 20

An approach to iteratively fitting model parameters with latent variable

Question 21

Variance

Answer 21

Measure of spread of a random variable

Question 22

Random Forests

Answer 22

Ensemble of decision trees

Question 23

RANSAC

Answer 23

Robust method to fit a model to data with outliers

Question 24

Curse of dimensionality

Answer 24

Issues in data sparsity

Question 25

Gini impurity

Answer 25

A definition of predictability

Question 26

Expectation Maximization

Answer 26

Algorithm to learn with latent variables

Question 27

Projection length

Answer 27

Similarity measure in subspace method

Question 28

k-fold cross validation

Answer 28

A technique for assessing a model while exploiting available data for training and testing

Question 29

Fisher’s criterion

Answer 29

An approach to find useful dimension for classification

Question 30

Occam’s razor

Answer 30

A principle to choose the simplest explanation

Question 31

Bagging

Answer 31

Bootstrap aggregating

Question 32

A priori probalitiy

Answer 32

Probability before observation

Question 33

Dimension of a subspace

Answer 33

The number of base vectors