Homepage › Solution manuals › Yaser Abu-Mostafa › Learning from Data › Exercise 1.1

Learning from Data

Exercise 1.1

Answers

1.

Input space $X$ : patient’s medical history, symptoms, personal health information etc.
Output space $Y$ : all possible diseases
target function $f : X \to Y$ : ideal formula to identify a patient’s problem
Data set: All available patients’ information and their corresponding correct problem diagnostic.

2.

Input space $X$ : handwritten digits (digitalized)
Output space $Y$ : 0-9 digits
target function $f : X \to Y$ : ideal formula match a handwritten digit to a correct digit
Data set: handwritten digits and their corresponding correct matches

3.

Input space $X$ : every information of an email, e.g. words
Output space $Y$ : yes/no
target function $f : X \to Y$ : ideal formula to identify whether an email is spam or not
Data set: Spam and non-spam emails that have been identified by human

4.

Input space $X$ : the price of electricity, temperature, day of the week
Output space $Y$ : electric load
target function $f : X \to Y$ : ideal function that gives exact electric load for a given price, temperature and day of the week.
Data set: historical electric load along with corresponding price, temperature and day of the week information.

5.

Input space $X$ : cat images
Output space $Y$ : types of cats
target function $f : X \to Y$ : ideal function that gives cat type according to the picture
Data set: pictures with cats that have been categorized to various types

niuers

2021-12-07 11:18

Comments