Lecture1-4Components of ML

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 as g.

Hypothesis set 
- can contain good and bad hypotheses
- up to  to pick the ‘best’ h as g

 is the ML algorithm.

Learning Model =  and 

Learning Flow 2