Task 3 Not as important

Question 1

How do we test mental representation theories ? (not as important)

Answer 1

• Psychological plausibility
• Neurological plausibility
• practical applicability
• Representational power
• Computational power

Question 2

What is meant by Representational power ?

Answer 2

• How much information a mental representation can express.

- What do the mental representations express?

Question 3

What is meant by Computational power ?

Answer 3

• Basically, what can you do with the mental representation in terms of high lvl thinking
• Can it be used for problem solving, (including planing decison making and explanation), learning and the use of langaug
• how efficient is ur computational procedure acting on the mental representation

Question 4

What is meant by practical applicability ? (not as important)

Answer 4

• Can the mental representation theory be used in fields of education and artifical intelligence

Question 5

What is meant by psychological plausibility ? (not as important)

Answer 5

• can the mental representation explain human thinking / undertsanding human cognition

Question 6

What is meant by neurological plausibility ? (not as important)

Answer 6

• Is the mental representation theory consistent with the results of neuroscientific experiments

Question 7

How do machine learn ?

Answer 7

• Via an optimization algorithm

Question 8

What does the optimization algorithm consist of ?

Answer 8

1. labeld data (numbers)

2. tunable parameters

Question 9

How to find the best possible output ?

Answer 9

• Via finding the perfect combination of parameters between all layers

Question 10

Ho do we find out the best possible parameters combination ?

Answer 10

1. Via randomly trying (bad)
2. Via the gradient method
   - Both try to find the local minimum of the coast function

Question 11

What does deep learning mean ?

Answer 11

• each layer can use the information extracted in the previous layer to build up a more complex representation of the data

Question 12

What is part of a parameter ?

Answer 12

• A parameter consist of weights and a Bias

Question 13

How does the gradient technique work ?

Answer 13

1. You take a randome parameter
2. Coast function tells u the parameter is incorrect
3. the coast function reduces the error via changing the bias and its weights

Question 14

How does the coast function work?

Answer 14

• The lower the coast function the better the parameter
• The higher the coast function the worst the parameter
• It is the way to say that the network makes an error

Question 15

How does the coast function now in which direction the weight should shift in order to become a better output ? (2 Dimensional)

Answer 15

• Identfying the slope of the randome used parameter
• shift to the left if slope is positive and shift the input to the right if the slope is negative
• Right = increase the parameter
• left = decrease the parameter
• If you do this over and over again you will approach a local minimum

Question 16

How does the coast function now in which direction the weight should shift in order to become a better output ? (3 Dimensional)

Answer 16

• identify the gradient of a function
• Make the gradient negative which will show you in which direction you will have to move your parameter
• Do this over and over again to adjust the parameters (gradient descent)
• Identify the local minimum of the coast function

Question 17

Why do we need to use the negative gradient in order to identfy the local minimum of the coast function ?

Answer 17

• The normal gradient = steepest highest increase

- The negative gradient shows decrease most quick

Question 18

What is meant by gradient descent ?

Answer 18

• Repeated process of following the negative gradient to adjust parameters
• Also identifies which weights matter more then other regarding the output

Question 19

What are the different approaches to mental representation ?

Answer 19

• logic, rules, concepts, images, connections (artificial neural networks)