第 72 / 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
4177One-Hot Plus Interaction Column CountA categorical variable has 5 levels. You one-hot encode it with a dropped baseline, keep one raw numeric feature x, and also create all interactions between x and the retained dummies. How many columns come out of this block in total?机器学习简单数值题未尝试面试订阅4178Winsorize-then-Standardize PipelineA raw daily return of 4.8% is winsorized to the range [-3%, 3%], then standardized using trailing mean 0.5% and trailing standard deviation 1.0%. What z-score feature results?机器学习简单数值题未尝试面试订阅4179Log1p Volume TransformA liquidity feature uses log1p(volume). If today's volume is 999999 shares, what transformed value do you store?机器学习简单数值题未尝试面试订阅4180Leakage-Safe Rolling Mean FeatureAt today's open you build a leakage-safe rolling-mean return feature from the last four completed daily returns: [1.0%, -2.0%, 0.5%, 1.5%]. What feature value do you use?机器学习简单数值题未尝试面试订阅4181Rolling Mean with a Forward WindowA candidate constructs a '5-day moving average' for day t using prices from t-2 through t+2. Is this valid for a predictive model meant to trade at the close of day t?机器学习中等derivation未尝试面试订阅4182Standardization Before the Train-Test SplitA pipeline standardizes each feature using the mean and standard deviation of the full dataset before creating the train/test split. Is that clean?机器学习中等derivation未尝试面试订阅4183Target Encoding Without Out-of-Fold LogicA categorical feature is target-encoded using the full sample average label for each category, and then those encodings are used inside cross-validation. Is that safe?机器学习中等derivation未尝试面试订阅4184Lag Feature with an As-Of TimestampA feature uses yesterday's close, but only if the data vendor timestamp shows that the value was available before today's decision time. Is that construction conceptually acceptable?机器学习中等derivation未尝试面试订阅4185Cross-Sectional Rank Built After Universe FilteringA cross-sectional signal ranks stocks only within the subset that survived a future-based liquidity filter. Is that a valid engineered feature for backtesting?机器学习中等derivation未尝试面试订阅4186Why Centering Helps Interaction FeaturesWhy do practitioners often center features before adding interaction terms to a linear model?机器学习中等essay未尝试面试订阅4187Why More Features Can Hurt Linear ModelsWhy can adding many plausible engineered features make a linear model worse rather than better?机器学习中等essay未尝试面试订阅4188Why Dummy Variable Traps Are More Than a Coding BugWhy is the dummy-variable trap more than just a harmless coding oversight?机器学习中等essay未尝试面试订阅4189Why Domain Features Still MatterIn an era of flexible models, why can careful domain-driven feature engineering still matter a lot for linear methods?机器学习中等essay未尝试面试订阅4190A Fast Sanity Check for Feature-Engineering AnswersWhat is a fast sanity check after solving a feature-engineering interview question?机器学习中等essay未尝试面试订阅4191Hyperplane Score and Margin 1Using the polynomial kernel K(x,z)=(x·z+1) 2, what is K((1,2),(2,0))?机器学习中等数值题未尝试面试订阅4192Hyperplane Score and Margin 2An RBF SVM uses K(x,z)=exp(-γ||x-z|| 2) with γ=0.5. If ||x-z|| 2=4, what kernel similarity is produced?机器学习中等数值题未尝试面试订阅4193Hyperplane Score and Margin 3For a soft-margin SVM, a training point has label y=1 and score f(x)=0.3. What hinge loss max(0,1-yf(x)) does it incur?机器学习中等数值题未尝试面试订阅4194Hyperplane Score and Margin 4A soft-margin SVM uses objective 0.5||w|| 2 + C Σ hinge i. If one point has hinge loss 1.2 and C=2, what penalty contribution does that point add to the objective?机器学习中等数值题未尝试面试订阅4195Hyperplane Score and Margin 5If a linear SVM has ||w||=5, what is the geometric margin width 2/||w||?机器学习中等数值题未尝试面试订阅4196Kernel Decision Function Evaluation 1A kernel SVM prediction uses two support vectors. Their signed contributions at a test point are +1.2 and -0.4, and the bias is -0.1. What score and predicted class result?机器学习中等数值题未尝试面试订阅