Recent questions tagged machine-learning

A binary classification problem could be solved with the two approaches described below:Approach 1 : Simple Logistic Regression (one neuron)Your output will be $\hat y = ...

Given input data $x \in \mathbb{R}^3$ , how many parameters do the following models have? Assume that linear components of all models have bias terms.Binary logistic regr...

Given a vector $a = [0.3, 0.3, 0.3]$, what is the result of using softmax activation on this vector?$[1, 1, 1]$$[\frac{1}{2}, \frac{1}{2}, \frac{1}{2}]$$[\frac{1}{3}, \fr...

James likes big cats. He has a collection of color images, each of which has been accurately labeled as being a leopard, jaguar, cheetah, or lion. He wishes to design a 4...

Given an $n$-dimensional vector $y$ :For any real number $c$, will the softmax of $c \cdot y$ b e the same as the softmax of $y$?(Please enter 1 for True and 0 for False)

What is the definition of $softmax(x_1,..,x_n)$? Recall that this function takes in a list of $n$ real numbers $x_1,..., x_n$ and outputs a list of $n$ real numbers. $sof...

We trained a three-way logistic regression and obtained weights$$w_a = (1,1, 0), w_b = (-1,1,1), w_c = (2,1,2)$$.What label would be given to the point $x = (0, 1, 1)$?A ...

Consider the probability of a label $y_n$ given an input vector $x_n$ and parameter vector $w$ for a binary classification problem, where $y_n \in \{+1, - 1\}$. The proba...

Given the probability model for a binary classification problem where the label $y_n \in \{ + 1 , -1\} $and the probability is defined as,$$P(y_n \mid x_n) = \begin{cases...

Consider the probability of a label yn given an input vector $X_n$ and parameter vector $w$ for a binary classification problem, where $y_n \in \{+1, - 1|\}$. The probabi...

Which of the following statements are true about logistic regression? 1. If the points are linearly separable (with a positive margin), then logistic regression (with ...

If no regularization is used and the training data is linearly separable, the optimal model parameters will tend towards positive or negative infinity.(Please enter 1 for...

Consider the following scatter plot of our two (standardized) features. Which of the following statements are true about an unregularized logistic regression model fit on...

Suppose you have a logistic regression model for spam detection, using a dataset with a binary outcome that indicates whether an email is spam (1) or not spam (0). The pr...

There are 5 balls in a bag. Each ball is either red or blue. Let $\theta$ (an integer) be the number of blue balls. We want to estimate $\theta$, so we draw 4 balls with ...

For the given statements about cross-entropy loss.$(i)$ The value of cross-entropy loss is always non-negative.$(ii)$ Minimizing the cross-entropy is equivalent to maximi...

$\text { Which function is not a convex function ? }$Sigmoid/Logistic function: $f(x)=1 /\left(1+e^{-x}\right)$.Linear function: $f(x)=3 x$.Square function: $f(x)=x^2$.f(...

For a binary classification problem, where the true class label $C \in\{0,1\}$, the cross-entropy loss is defined based on the predicted probability $P_C$ that a data poi...

In this question, assume that we are using the logistic regression model $\hat{y}=\sigma\left(x^T \theta\right)$.Suppose we want to modify cross-entropy loss to penalize ...

In this question, assume that we are using the logistic regression model $\hat{y}=\sigma\left(x^T \theta\right)$.Suppose we want to modify cross-entropy loss to penalize ...