Recent questions tagged machine-learning

Consider the following binary classification dataset,where circles denote the positive class and squares the negative class:Which (if any) of the decision boundaries coul...

Given the following training points for classification, where each point is represented a $(x_1 , x_2 , y )$：$$(1,0, +)$$ $$(1,1, +)$$ $$(0,1, - )$$Perform one pass of th...

In terms of the bias-variance decomposition, a 1-nearest neighbor classifier has _____than a 3 -nearest neighbor classifier.higher variance higher biaslower variance lowe...

How does the bias-variance decomposition of a ridge regression estimator compare with that of ordinary least squares regression? (Select one.)Ridge has larger bias, large...

Consider the error decomposition for a least squares regression model$\mathbf{E}_{x, y, D}\left[(h(x ; D)-y)^2\right]=\mathbf{E}_{x, D}\left[(h(x ; D)-\bar{h}(x))^2\right...

In the bias-variance decomposition, our model's risk was defined as $\mathbb{E}\left[\left(Y-f_{\hat{\theta}}(x)\right)^2\right]$. Which of the following is equal to $\ma...

If we assume that the data is being generated from a source with no noise $y=w^T x$, choosing a classifier $f^*(x)=w^{* T} x$ using OLS will make which of the following q...

Which of the following is TRUE about the bias-variance decomposition of test error? Expected test error is equal to bias $+v a r^2+$ noiseNoise can be reduced by using re...

Matthew decides to try a new method. He first comes up with a model $h(x)$, which has a bias of $B$ and a variance of $V$ on the original dataset. Next, he builds a new m...

We broke the least-squares error into three separate terms:$$\mathbf{E}\left[\left(y-f_\theta(x)\right)^2\right]=\mathbf{E}\left[(y-h(x))^2\right]+\mathbf{E}\left[\left(h...

Assuming a regularization penalty of the form $\lambda R(\theta)$. Complete the following illustration. Note that the x -axis is the regularization parameter $\lambda$ an...

Your team would like to train a machine learning model in order to predict the next YouTube video that a user will click on based on $m$ features for each of the previous...

The expected squared error can be decomposed into several important terms:$$\left[\left(Y-f_{\hat{\theta}}(x)\right)^2\right]=\sigma^2+\left(h(x)-\mathbb{E}\left[f_{\hat{...

The data model, $y_i=f\left(X_i\right)+\epsilon_i$, that justifies the least-squares cost function in regression. The statistical assumptions of this model are, for all $...

Which of the following can impact our model variance ? Select all that apply.The regularization coefficient $\lambda$.The choice of features to include in our design matr...

Consider a quintic (degree 5) model $\hat{y}=\hat{q}_0+\hat{q}_1 x+\ldots+\hat{q}_5 x^5$, where $\hat{q}_j$ is the MSE parameter estimate for feature $j$ of this quintic ...

Consider a quintic (degree 5) model $\hat{y}=\hat{q}_0+\hat{q}_1 x+\ldots+\hat{q}_5 x^5$, where $\hat{q}_j$ is the MSE parameter estimate for feature $j$ of this quintic ...

Ridge regressionreduces variance at the expense of bias.adds an $l_1$ penalty norm to the cost function.often sets many of the weights to 0 when the regularization parame...

Reducing the regularization of a model would typically . . .Decrease its bias and increase its variance.Decrease its bias and Decrease its variance.Increase its bias and ...

Which of the following is most indicative of a model overfitting?High bias, low varianceLow bias, high varianceLow bias, low varianceNone