第 17 / 88 页
非代码面试题
显示 20 / 1751 道匹配题目
答题状态：未尝试未正确已正确
ID题目领域难度题型进度权限
2465Why Nested Validation ExistsIf the same validation set is repeatedly used for model family choice, feature engineering, and threshold tuning, why is a second outer holdout or nested procedure conceptually necessary?机器学习困难essay未尝试面试订阅2470Rare Category Thresholding After Seeing Test CompositionSuppose you choose the minimum frequency for keeping a category only after inspecting how many rare categories appear in the test set. Why is that already a contaminated design choice?机器学习困难essay未尝试面试订阅2471Slope From Centered Sufficient Statistics 1In simple OLS with an intercept, if the centered sufficient statistics satisfy sum i (x i-xbar)(y i-ybar)=S xy and sum i (x i-xbar) 2=S xx, derive beta hat.机器学习简单derivation未尝试免费2472Intercept From Sample Means and Slope 2Derive the OLS intercept in simple regression with an intercept once beta hat is known.机器学习简单derivation未尝试免费2473Scaling One Feature Rescales Its Coefficient 3If a feature x is replaced by x new = c x in an OLS model with an intercept, how does its fitted coefficient change when all fitted values are kept identical?机器学习中等derivation未尝试面试订阅2475Why Duplicate Features Cause Non-Unique Coefficients 5Why do two perfectly duplicated features make the OLS coefficient vector non-unique even though fitted predictions can stay unique?机器学习困难essay未尝试面试订阅2477Why Centering Can Simplify OLS Algebra 7Why does centering features and targets often make OLS derivations cleaner when an intercept is present?机器学习中等essay未尝试免费2478Residual Orthogonality to Features 8Why must the OLS residual vector be orthogonal to every column of the design matrix at the optimum?机器学习中等derivation未尝试面试订阅2480Orthogonal Features Give Coordinatewise Coefficients 9Suppose two features x1 and x2 are centered and orthogonal. Derive the OLS coefficients in terms of x1 T y, x2 T y, ||x1|| 2, and ||x2|| 2.机器学习困难derivation未尝试面试订阅2483Why Centering Leaves Slopes Unchanged 13Why does centering x and y leave the fitted slope unchanged in simple OLS with an intercept?机器学习中等derivation未尝试面试订阅2484Response Scaling 14If every target is multiplied by c, what happens to the OLS coefficient vector and intercept?机器学习困难derivation未尝试面试订阅2485Why Gradient Descent and Closed Form Agree 15Why do exact gradient descent convergence and the normal-equation solution agree for OLS?机器学习困难derivation未尝试面试订阅2487Prediction Invariance Under Equivalent Parameterizations 16Why can two different coefficient vectors produce exactly the same OLS predictions when the design is rank-deficient?机器学习中等derivation未尝试面试订阅2489A Constant Shift in One Feature 19If a feature x is replaced by x+k in a regression that already includes an intercept, what happens to the slope on x and the intercept?机器学习中等derivation未尝试面试订阅2490Why OLS Can Still Predict Well Under Misspecification 20Why can OLS remain a useful predictor even when the true data-generating process is not exactly linear?机器学习困难essay未尝试面试订阅2495When OLS Predictions Are Unique 25Even if the coefficient vector is not unique, why is the OLS fitted prediction X beta hat still unique?机器学习困难derivation未尝试面试订阅2497Why Ridge Shrinks but Rarely Zeros 2Why does ridge typically shrink coefficients continuously toward zero rather than setting many of them exactly to zero?机器学习简单essay未尝试免费2498Lasso Zero Threshold in an Orthogonal Coordinate 3In an orthogonal one-feature problem with x T x = d and score z = x T y, for what lambdas does the lasso coefficient become exactly zero?机器学习中等derivation未尝试免费2499Soft-Thresholded Lasso Coefficient 4In an orthogonal one-feature problem with x T x = d and x T y = z > 0, derive the lasso coefficient when 0 < lambda < z.机器学习中等derivation未尝试面试订阅2500Equivalent Lambda for a Target Ridge Shrinkage Ratio 5In an orthogonal coordinate, ridge shrinks beta ols by the factor d/(d+lambda). What lambda yields a shrinkage ratio r in (0,1)?机器学习困难derivation未尝试面试订阅