Problem Solving Under Uncertainty Flashcards
what’s dead reckoning?
the process of using information about direction, speed, and elapsed time to calculate the new location.
what does odometry refer to?
he use of motion sensors, such as measuring wheel rotation
what’s the difference between deductive and inductive inferences?
deductive = is all about going from true statements to other true statements using rules of logic. we are certain about our knowledge. We infer facts about the world.
-all Italian cities have a church
-Senago is an Italian city
-thus Senago has a church
inductive=about going from specific observations to informed guesses, or conjectures. Uncertain about our knowledge.
We make conjectures about the world.
-i observe a bunch of cities in Italy
-they all have a church
-i conclude all italina cities have a church
what’s an observation?
it describes a set of inputs and an output
- the input x is: a feature, attribute, or input variable
- output (y) is: output variable
series of observations: k
why is making predictions difficult in the weather example?
- There is likely to be noise in the data.
- There are yearly variations in temperature.
- We want to capture what is systematic in London’s temperature.
- We want to ignore what is accidental in London’s temperature.
what does systematic mean?
that something is likely to be observed again.
what does accidental mean?
tat something is not likely to b observed again
what does a second-degree polynomial look like? what are its parameters and expression?
expression :f (x) = a0 + a1x + a2x2
Appearance: quadratic curve ( U )
parameters: 3
What’s the difference between fitting and predicting?
-fitting refers to how well the trained model describes the observations.
- are interested in how well this trained
model predicts new observations.
when fitting what does it mean if the polynomial degree is high?
it means theres less errors
