• retagged by
5,413 views
7 7 votes

​​​​​​Consider the dataset with six datapoints: $\left\{\left(\text{x}_{1}, \text{y}_{1}\right),\left(\text{x}_{2}, \text{y}_{2}\right), \ldots,\left(\text{x}_{6}, \text{y}_{6}\right)\right\}$, where $x_{1}=\left[\begin{array}{l}1 \\ 0\end{array}\right], x_{2}=\left[\begin{array}{l}0 \\ 1\end{array}\right], x_{3}=\left[\begin{array}{c}0 \\ -1\end{array}\right], x_{4}=\left[\begin{array}{c}-1 \\ 0\end{array}\right], x_{5}=\left[\begin{array}{l}2 \\ 2\end{array}\right], x_{6}=\left[\begin{array}{l}-2 \\ -2\end{array}\right]$ and the labels are given by $\text{y}_{1}=\text{y}_{2}=\text{y}_{5}=1$, and $\text{y}_{3}=\text{y}_{4}=\text{y}_{6}=-1$. A hard margin linear support vector machine is trained on the above dataset.


Which ONE of the following sets is a possible set of support vectors?

  1. $\left\{x_{1}, x_{2}, x_{5}\right\}$
  2. $\left\{x_{3}, x_{4}, x_{5}\right\}$
  3. $\left\{x_{4}, x_{5}\right\}$
  4. $\left\{x_{1}, x_{2}, x_{3}, x_{4}\right\}$

 

2 Answers

10 10 votes

Here, $x_1,x_2,x_3,x_4$ are the support vectors.

    

  

In Support Vector Machine(SVM) statistical learning method, we have to maximize the margin. Support vectors (difficult points)  are those data points that the margin pushes up against the decision boundary.

   

Hard-Margin SVM is for linearly separable data otherwise we use the soft-margin SVM for non-linearly separable data using the kernel trick.

   

If the Largrangian Multiplier for the optimization of the objective function is $\alpha$ then $\alpha \neq 0$ for the support vectors and for rest of the points, it is zero. So, we can verify it by solving the convex optimization problem for the SVM and find the Lagrangian Multiplier and check whether it is non-zero or not.    

1 1 vote

 

 

This can be done mentally. Any point with a large co-ordinate like $[2,2]$ or $[-2,-2]$ must be eliminated, as it is too far to be a Support Vector. Hence, we omit $x_5$ and $x_6$. Options A, B, C all have $x_5$. Only Option D doesn't contain $x_5$ or $x_6$. 

Correct Answer: Option D.

• edited by
Answer:
Position:
Show:

Related questions

6 6 votes
1 1 answer
4.4k
4.4k views
Arjun asked Feb 16, 2024
4,396 views
​​​​​Match the items in Column $1$ with the items in Column $2$ in the following table:\[\begin{array}{|l|l|}\hline\text{Column 1} & \text{Column 2} \\\hline\text{(p) Pri...
3 3 votes
3 3 answers
6.5k
6.5k views
Arjun asked Feb 16, 2024
6,450 views
​​​​​​Euclidean distance based $k$-means clustering algorithm was run on a dataset of $100$ points with $k=3$. If the points $\left[\begin{array}{l}1 \\ 1\end{array}\righ...
8 8 votes
2 2 answers
8.0k
8.0k views
Arjun asked Feb 16, 2024
8,009 views
​​​​​​Given a dataset with $K$ binary-valued attributes (where $K>2$ ) for a two-class classification task, the number of parameters to be estimated for learning a naïve ...
5 5 votes
3 3 answers
4.8k
4.8k views
Arjun asked Feb 16, 2024
4,764 views
​​​​​​For any binary classification dataset, let $S_{B} \in \mathbb{R}^{d \times d}$ and $S_{W} \in \mathbb{R}^{d \times d}$ be the between-class and within-class scatter...