代写COMPSCI5100 Machine Learning for Data Scientists 19-20代写Web开发

DEGREES OF MSc, MSci, MEng, BEng, BSc,MA and MA (Social Sciences)

COMPSCI5100

Machine Learning for Data Scientists

1. A polynomial regression model is defined as:

 

(a)  When applying this model to the Olympic data, where xn ∈ 1896) . . . )2008 , we always 

rescale x, e.g.  Explain why is this rescaling necessary.         [4 marks]

(b)  Write down the regression model if the following Radial basis function (RBF) with a range of different position parameter μ is applied to xn .

 

[3 marks]

(c)  Give an example of nonlinear regression model. Also explain why it is nonlinear .           [2 marks]

(d)  Explain why polynomial regression could suffer from outliers .    [3 marks]

(e)  L2 Regularised regression can be used to deal with this problem . Use a contour plot (Assuming the dimension of the parameter is 2) of parameters and the loss function to explain why.   [8 marks]

2.  Classification question

(a)  Classification and regression are both supervised learning problems . Describe a way to turn a regression problem into a classification problem? (Please state the differences between the two.)         [3 marks]

(b)  The receiver operating characteristic (ROC) curve is a standard way to visualize the performance of classifiers .  Outline how to draw a ROC curve              [3 marks]

(c) For the classifier outputs in the table below, provide a value for the missing output (labeled ‘?’) that would:

(i)        Give an AUC equal to 1 ,                 [2 marks]

(ii)       Give an AUC less than 1 ,                  [2 marks]

(iii)      Now assuming you can change any output of the six data points, give

an example of the outputs , such that the AUC is 0.5.             [3 marks]

(d) Use a diagram (some data in 2D) to describe how linear Support Vector Machines (SVMs) operate (how they make classification decisions, what and how parameters have to be set, what data needs to be stored etc) .       [4 marks]

(e) Logistic regression uses the sigmoid function to make classification decision .

Now, we have defined the following probability with a sigmoid function .

 

Write down the corresponding likelihood function for nth data points, p(tn Iw) xn).         [3 marks]

3. Unsupervised learning question

Consider using the K-means algorithm to perform. clustering on the following data .

We want to cluster data in the outer ring in one cluster and the data in the inner circle as a different cluster .

(a)  Outline what would happen if we directly apply K-means with Euclidian distance to

this data . Can it achieve the clustering objective? How will it split/group the data?               [2 marks]

(b)  An alternative approach is to use Kernel K-means. Explain how the kernel could help in this dataset.   [3 marks]

(c)  Which one of the following statements about kernel is NOT correct?

A.  One could use the RBF kernel, K(xn) xi ) =  exp(一Y(xn  一 xi )T (xn  一 xi )), to achieve the clustering target.

B.  The RBF kernel projects the data onto an infinite dimensional space . The free parameter Y can be estimated with cross-validation .

C.  One could use the linear kernel , K(xn) xi ) =  xn(T)xi , to achieve the clustering target.

D.  The linear kernel projects the data onto itself, and there is no free parameter to tune .           [2 marks]

(d)  Write some pseudo-code to perform. K-means .              [5 marks]

(e)  Outline how and why cross-validation can be used to select the number of clusters K. [8 marks]




热门主题

课程名

mktg2509 csci 2600 38170 lng302 csse3010 phas3226 77938 arch1162 engn4536/engn6536 acx5903 comp151101 phl245 cse12 comp9312 stat3016/6016 phas0038 comp2140 6qqmb312 xjco3011 rest0005 ematm0051 5qqmn219 lubs5062m eee8155 cege0100 eap033 artd1109 mat246 etc3430 ecmm462 mis102 inft6800 ddes9903 comp6521 comp9517 comp3331/9331 comp4337 comp6008 comp9414 bu.231.790.81 man00150m csb352h math1041 eengm4100 isys1002 08 6057cem mktg3504 mthm036 mtrx1701 mth3241 eeee3086 cmp-7038b cmp-7000a ints4010 econ2151 infs5710 fins5516 fin3309 fins5510 gsoe9340 math2007 math2036 soee5010 mark3088 infs3605 elec9714 comp2271 ma214 comp2211 infs3604 600426 sit254 acct3091 bbt405 msin0116 com107/com113 mark5826 sit120 comp9021 eco2101 eeen40700 cs253 ece3114 ecmm447 chns3000 math377 itd102 comp9444 comp(2041|9044) econ0060 econ7230 mgt001371 ecs-323 cs6250 mgdi60012 mdia2012 comm221001 comm5000 ma1008 engl642 econ241 com333 math367 mis201 nbs-7041x meek16104 econ2003 comm1190 mbas902 comp-1027 dpst1091 comp7315 eppd1033 m06 ee3025 msci231 bb113/bbs1063 fc709 comp3425 comp9417 econ42915 cb9101 math1102e chme0017 fc307 mkt60104 5522usst litr1-uc6201.200 ee1102 cosc2803 math39512 omp9727 int2067/int5051 bsb151 mgt253 fc021 babs2202 mis2002s phya21 18-213 cege0012 mdia1002 math38032 mech5125 07 cisc102 mgx3110 cs240 11175 fin3020s eco3420 ictten622 comp9727 cpt111 de114102d mgm320h5s bafi1019 math21112 efim20036 mn-3503 fins5568 110.807 bcpm000028 info6030 bma0092 bcpm0054 math20212 ce335 cs365 cenv6141 ftec5580 math2010 ec3450 comm1170 ecmt1010 csci-ua.0480-003 econ12-200 ib3960 ectb60h3f cs247—assignment tk3163 ics3u ib3j80 comp20008 comp9334 eppd1063 acct2343 cct109 isys1055/3412 math350-real math2014 eec180 stat141b econ2101 msinm014/msing014/msing014b fit2004 comp643 bu1002 cm2030
联系我们
EMail: 99515681@qq.com
QQ: 99515681
留学生作业帮-留学生的知心伴侣!
工作时间:08:00-21:00
python代写
微信客服:codinghelp
站长地图