Exercise 1 - Basic Concepts in Machine Learning

Question 1

Learning

Answer 1

Learning is the acquisition of new information or knowledge or the process to acquire knowledge or skill by systematic study or by trial and error.

Question 2

Google Translate ML?

Answer 2

Yes

Question 3

calls with Google to restaurant ML based?

Answer 3

Yes. Only work in small domain

Question 4

WildCat (The World’s Fastest Quadruped RobotI) ML based?

Answer 4

No

Question 5

ML definition

Answer 5

• field of study that gives computers the ability to learn without being explicitly programmed

Question 6

Four components of a ML system

Answer 6

• Dataset
• Model
• Objective Function
• Algorithm

Question 7

Is ML a prerequisite for the implementation of cognitive functions in artificial cognitive systems?

Answer 7

• yes

Question 8

four cognitive functions in artificial cognitive systems

Answer 8

• Learning and development
• Memory, knowledge, and internal simulation
• Perception
• Autonomy

Question 9

Learning and Development in the context of artificial cognitive systems

Answer 9

• modelling and implementation of biological learning mechanisms (operant conditioning, implicit learning, explicit learning, perception)

Question 10

implicit learning

Answer 10

learning without conscious operations

Question 11

explicit learning

Answer 11

conscious operation where the individual works and tests hypothesis in a search for structure

Question 12

How is perception implemented in ML?

Answer 12

• e.g. unsupervised learning of visual features

Question 13

What is autonomy?

Answer 13

dynamic adaption to changes in the environment

Question 14

For which environments is ML useful?

Answer 14

For artificial cognitive systems that are situated in complex dynamic environments.

Question 15

Why is it not possible to program everything in advance in complex dynamic environments?

Answer 15

• environment is changing continuously
• dynamics too complex to be modeled explicitly (faces)
• system itself is subject to change (through growth, aging etc. )

Question 16

feature engineering

Answer 16

• selection of the right features. Features must contain the information required for predicitions

Question 17

inductive learning

Answer 17

• specific examples -> general rule

Question 18

Definition of ML task

Answer 18

Train model M in a hypothesis space H using a learning algorithm A so that M minimizes loss L for dataset S. This type of learning is inductive learning.

-> The choice of H and L depends heavily on the properties of S

Question 19

Mixed data

Answer 19

labeled and unlabeled data

Question 20

Dynamic data

Answer 20

Zeitreihen

Question 21

Do natural datasets have specific structural featues?

Answer 21

Yes.

Question 22

Semi-supervised learning

Answer 22

• Mixed data: labeled and unlabeled training samples

- a priori assumptions on input data required

Question 23

reinforcement learning

Answer 23

-Dynamic environment: interaction with the environment
- reward signal encodes feedback for the policy
(trail and error)

Question 24

Types of ML

Answer 24

Unsupervised learning
Semi-supervised learning
Reinforcement learning
Supervised learning

Question 25

What is the choice of a learning paradigm motivated by?

Answer 25

The type of data available