# Tutor profile: Jordan A.

## Questions

### Subject: MATLAB

What is the difference between M-file and MEX files in Matlab?

M files: They are just a plain ASCII text that is interpreted at run time. They are like sub-programs stored in text files with .m extensions and are called M-files. For most of MatLab, development M-files are used. MEX files: They are basically native C or C++ files which are linked directly into the MatLab application at runtime. MEX files have efficiency to crash the MatLab application.

### Subject: Machine Learning

What does "linear machine" mean in the context of classification? If using Bayesian decision theory and assuming a Gaussian pdf, what additional assumptions are needed in order to result in a linear machine? Give two examples.

A linear machine means that the decision boundary is a line in 2D or a hyperplane in higher dimensions. The first assumption is that the features are independent. The second assumption is that the classes have the same variance (ex: minimum Euclidean distance classifier) or the different classes have the same covariance matrix (ex: minimum Mahalanobis distance classifier).

### Subject: Calculus

Two people are $$50$$ feet apart. One of them starts walking north at a rate so that the angle between them is changing at a constant rate of 0.01 rad/min. At what rate is the distance between the two people changing when $$\theta=0.5$$ radians?

Let the distance between the two people be $$x$$. We can relate the two distance quantities with the following trig formulas $(\cos\theta=\frac{50}{x} \textrm{ and } \sec\theta=\frac{x}{50}.$) We want to find $$x',$$ so we use the second equation. Note we could also find $$x$$ using the first equation. Thus, we differentiate the second equation to get $(\sec\theta\tan\theta \theta' =\frac{x'}{50}$). We want to find $$x'$$ and know that $$\theta=0.5$$ and $$\theta'=0.01$$, therefore solving for $$x'$$ we get $( x'=(50)(0.01)\sec(0.5)\tan(0.5)=0.31125 \textrm{ ft/sec} .$)

## Contact tutor

needs and Jordan will reply soon.