# Target function f x y ideal credit approval formula

• Notes
• 50
• 100% (1) 1 out of 1 people found this document helpful

Course Hero uses AI to attempt to automatically extract content from documents to surface to you and others so you can study better, e.g., in search results, to enrich docs, and more. This preview shows page 35 - 42 out of 50 pages.

target function:f:X → Y(ideal credit approval formula)datatraining examples:D={(x1,y1),(x2,y2), · · · ,(xN,yN)}(historical records in bank)hypothesisskillwith hopefullygood performance:g:X → Y(‘learned’ formula to be used){(xn,yn)}fromfMLg
Some Key WordsHypothesis:A hypothesis is a certain function that we believe (or hope) is similar to the truefunction, thetarget functionthat we want to model. In context of email spam classification, it wouldbe therulewe came up with that allows us to separate spam from non-spam emails. It is arelationship (pattern) between the input and output values. Lets say that this the functiony=f(x).However, f(.) is unknown function to us.so machine learning algorithms try to guess a``hypothesis'' function h(x) that approximates the unknown f(.). The set of all possible hypotheses isknown as theHypothesis set H(.).The goal in the learning process is to find the final hypothesisgthatbest approximates the unknown target functionf. Different machine learning models have differenthypothesis sets,Learning Algorithm:The ML goal is to find or approximate thetarget function, and thelearning algorithmis a set of instructionsthat tries tomodelthe target function using our trainingdataset. A learning algorithm comes with ahypothesis space, the set of possible hypotheses it cancome up with in order to model the unknown target function by formulating thefinal hypothesisSJSU CMPE297-02 Machine Learning36
Learning Flowunknown target functionf:X → Y(ideal credit approval formula)training examplesD: (x1, y1), · · · ,(xN, yN)(historical records in bank)learningalgorithmAfinal hypothesisgf(‘learned’ formula to be used)
Learning Flowunknown target functionf:X → Y(ideal credit approval formula)training examplesD: (x1, y1), · · · ,(xN, yN)(historical records in bank)learningalgorithmAfinal hypothesisgf(‘learned’ formula to be used)targetfunknown(i.e. no programmable definition)hypothesisghopefullyfbut possiblydifferentfromf(perfection ‘impossible’ whenfunknown)
Learning Flowunknown target functionf:X → Y(ideal credit approval formula)training examplesD: (x1, y1), · · · ,(xN, yN)(historical records in bank)learningalgorithmAfinal hypothesisgf(‘learned’ formula to be used)targetfunknown(i.e. no programmable definition)hypothesisghopefullyfbut possiblydifferentfromf(perfection ‘impossible’ whenfunknown)What doesglook like?
The Learning Modeltraining examplesD: (x1, y1), · · · ,(xN, yN)(historical records in bank)learningalgorithmAfinal hypothesisgf(‘learned’ formula to be used)hypothesis setH(set of candidate formula)assumeg∈ H={hk}, i.e. approving ifh1: annual salary > \$30,000h2: debt > \$50,000 (really?)h3: year in job2 (really?)hypothesis setH:can containgood or bad hypotheses
The Learning Modeltraining examplesD: (x1, y1), · · · ,(xN, yN)(historical records in bank)learningalgorithmAfinal hypothesisgf(‘learned’ formula to be used)hypothesis setH(set of candidate formula)assumeg∈ H={hk}, i.e. approving ifh1: annual salary > \$30,000h2: debt > \$50,000 (really?)h3

Course Hero member to access this document

Course Hero member to access this document

End of preview. Want to read all 50 pages?