Machine Learning Basics

Question 1

A Model

Answer 1

A trained algorithm which is used to identify patterns in your data and does not require explicit manually set rules.

Question 2

Weight

Answer 2

How much does that feature affect the accuracy of the prediction?

Question 3

Sample Formula

Answer 3

(a0x0) + (a1x1) + (a2x2) + … = ?
> 1 = recommendation

Question 4

Supervised Learning

Answer 4

Learn by identifying patterns in data that’s already been labeled.

Question 5

Types of Supervised Learning

Answer 5

1) Classification
1a) Binary classification (Yes, No)
1b) Multiclass classification (IT Support, Returns, Accounting)
2) Regression (x -> 113 to 127)

Question 6

Regression

Answer 6

Continuous value like an integer
I.E. Stock Price

Question 7

Unsupervised

Answer 7

The machine has to uncover and create the labels itself.

Example: Clustering data.

Question 8

Unsupervised Use Case

Answer 8

Anomalies.

Are anomalies the result of outliers or indicators of hardware failure.

Question 9

Reinforcement Learning

Answer 9

Action & Reward/Penalty

Question 10

Deep Learning

Answer 10

Subset of Machine Learning

Artificial neural networks

Question 11

Chess examples of AI/ML/DL

Answer 11

AI: Machines that can play chess based on rules

ML: Machines learn to play chess from analyzing past chess games played by humans.

DL: Machines that can learn to play chess by playing against themselves

Question 12

Artificial Neural Networks

Answer 12

The way in which tasks are learned.

Question 13

ImageNet Large Scale Visual Recognition Challenge

Answer 13

https://www.image-net.org/challenges/LSVRC/

Computer defeated human in 2015.