第 3 / 5 页
非代码面试题
显示 20 / 89 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
2654Expected Number of Times a Point Is Validated in Repeated k-Fold CVIn R repeats of ordinary k-fold CV, each point appears in exactly one validation fold per repeat. Derive the number of validation appearances of one point across all repeats.机器学习中等derivation未尝试面试订阅2655Why Expanding Windows Can Beat Rolling Windows Under Sparse DataWhy might an expanding-window CV design be preferable to a rolling-window design when the series is short and drift is present but not violent?机器学习困难essay未尝试面试订阅2656Why Random Row CV Breaks With Overlapping Label HorizonsWhy can ordinary random row cross-validation severely overstate performance when each label depends on the next 5 trading days and adjacent rows overlap in those horizons?机器学习简单essay未尝试免费2657Why Fold-to-Fold Variation Is Not a Standard Error of DeploymentWhy is the standard deviation of fold scores not automatically the standard error of future production performance?机器学习简单essay未尝试免费2658Repeated Fold Inspection as a Tuning ChannelA researcher keeps examining which dates underperform in each fold and then adjusts features accordingly. Why is this still overfitting even if no formal hyperparameter optimizer is used?机器学习中等essay未尝试面试订阅2659Train-Set Size After Purging One Side of a FoldA dataset has 500 time-ordered observations. One validation block uses observations 301 through 350. If a 10-observation purge is applied immediately before the validation block and nowhere else, how many observations remain eligible for training?机器学习困难数值题未尝试面试订阅2660Why Rare-Event Stratification MattersWhy can ordinary random folds become misleading in a rare-event problem even when the data are IID?机器学习困难essay未尝试面试订阅2661Why Time-Series CV Is About Information Availability, Not Calendar PurityWhy is the real principle in time-series CV 'never train on information from the future' rather than 'always use a particular fold geometry'?机器学习简单essay未尝试免费2662Why Overlapping Validation Windows Complicate Score AggregationWhy should a practitioner be careful when averaging performance over overlapping validation windows?机器学习中等essay未尝试面试订阅2663Why Comparing CV Scores Across Different Fold Rules Can MisleadWhy is it dangerous to compare one model's score from random k-fold CV with another model's score from grouped or blocked CV?机器学习中等essay未尝试面试订阅2664How Many Distinct Hyperparameter Winners Can Outer CV ProduceA nested CV uses 7 outer folds and selects exactly one hyperparameter setting inside each outer fold. What is the maximum possible number of distinct winning hyperparameter settings across outer folds?机器学习困难derivation未尝试面试订阅2665Why Tiny Folds Can Exaggerate RegularizationWhy can a very small training fold make heavily regularized models look better than they would on the full training set?机器学习困难essay未尝试面试订阅2667Why Class Stratification Is Not Enough for Repeated EntitiesWhy can class-stratified cross-validation still fail badly when the same issuer appears many times and issuer identity carries predictive information?机器学习简单essay未尝试免费2668Why Embargo Matters Even With Backward-Looking FeaturesSuppose features use only past prices, yet labels depend on future returns over an event window. Why can an embargo still be necessary around the validation block?机器学习中等essay未尝试面试订阅2669Why Purging and Embargo Solve Different ProblemsWhy is purging not the same thing as embargoing in time-series validation?机器学习中等essay未尝试面试订阅2670Why the Best CV Design Depends on the Deployment UnitWhy should the fold rule mirror the unit on which the model will actually generalize in production?机器学习困难essay未尝试面试订阅2696False Strategy Surviving Two Independent Research GatesA desk tries 80 genuinely null strategy ideas. A strategy is kept only if it passes an in-sample screen at 10% and then a fresh out-of-sample confirmation at 5%, with the two tests treated as independent under the null. What is the probability at least one null idea survives both gates?机器学习简单derivation未尝试面试订阅2697False Winner Probability After Clustering 240 Variants into 24 FamiliesA researcher generates 240 heavily correlated strategy variants but argues they amount to only 24 effectively independent families. If the desk still flags any family with p-value below 8%, what is the approximate probability of at least one false family-level winner under the null?机器学习简单数值题未尝试面试订阅2698Single-Test Significance Needed to Cap Familywise False Discovery at 10%A desk plans to try 60 effectively independent strategy ideas. What single-test significance level alpha keeps the probability of at least one false winner at exactly 10% under the independence approximation?机器学习中等数值题未尝试面试订阅2702Chance the Best Null t-Statistic Exceeds 2.4 Across 50 VariantsSuppose 50 genuinely null standardized t-statistics are approximately independent N(0,1). What is the probability the largest of them exceeds 2.4?机器学习中等数值题未尝试面试订阅