COSC2673/COSC2793 | Machine Learning
Tutorial | Week 02
1. What are the three main paradigms of Machine Learning? (a) Briefly describe each paradigm
(b) How does each paradigm differentiate?
(c) Name one application of each paradigm
2. Consider the following problem:
“Rebecca runs a non-profit taxi service that delivers meals and provides transportation for the elderly, mobility impaired and disabled. Her fleet of delivery and transportation vehicles, which includes vans, cars, motorbikes and bicycles, have GPS fitted to them, as well as the drivers have mobile phones issued by the organisation to help them navigate and to show them job/task schedules and locations. The organisation also has some information about clients and their history of job/task requests. Currently, when there is a request, the closest driver is allocated the job, and if it is a delivery job they will have to return to base first. This has led to some drivers driving long distances, untimely arrivals and sometimes unhappy clients and drivers. Rebecca has heard about ‘machine learning’ and wants to investigate how it can improve her service.”
Discuss possible strategies to help Rebecca using machine learning. Focus on what the learning problems are, what type of machine learning can be used to solve for each. In addition, discuss what might be some potential issues, including ethical and societal?
3. Consider the following equation/function:
Plot the function.
4. Consider the following equations: (a) J1 = ni=1 2i
(b) J2 = ni=1 ix
(c) J3 = ni=1 xi
y = ax + b
Describe what each summation equation means.
5. Consider the equation from Question 3. Derive its gradient.
6. Consider the following equation
y = (ax − b)2
Derive its gradient, using the chain rule.
7. Consider the following row vectors and matrix (a) x = [−1,4,5]
(b) y=[0,−3,10]
−2 3 4 (c)M= 1 2 0
Calculate the following: (a) xy
(b) xT
(c) xyT
(d) xTy (e) Mx (f) MxT
1