Components of ML: Metaphor using credit approval
The bank has the applicant information, and there is unknown pattern to be learned, which is used to solve the final question:
whether approving credit card good is good for bank?
Basic Notations
- input: x∈ (custorer application)
- output: y∈ (good/bad after approving credit card)
- unknown pattern to be learned
⟺
target function:
f :→ (ideal credit approval formula) - data ⟺ training examples: ={(x1,y1),(x2,y2),...,(xN,yN)} (historical records in bank)
- hypothesis
⟺
skill with hopefully good performance:
g :→ (‘learned’ formula to be used)
{(xN,yN)} from f→ ML →g
f is unkown and
g is learned!! Hopefully g≈f , and perfection is impossible for unknown f .
Learning Flow1
There will be many
h in the hypothesis set , and then we hope to select one good h asg .
Hypothesis set
- can contain good and bad hypotheses
- up to
to pick the ‘best’
h
as
is the ML algorithm.
Learning Model = and
Learning Flow 2