第 29 / 37 页
非代码面试题
显示 20 / 738 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
5039EM Mixture Diagnostic 24Why can k-means and a Gaussian mixture with shared spherical covariances give similar clusters, yet still disagree on borderline points?机器学习困难essay未尝试面试订阅5040EM Mixture Diagnostic 25Why should a desk be careful before interpreting mixture components as economic regimes rather than as a flexible density fit?机器学习困难essay未尝试面试订阅5046Recover Discordant Counts From McNemar Summary 6Two classifiers are compared on the same test set. The total number of discordant cases is b+c=16, model A is better so b>c, and the continuity-corrected McNemar statistic is 3.0625. What discordant counts (b,c) are implied?机器学习中等数值题未尝试面试订阅5051Cost-Sensitive Deployment Choice 11Model A makes 8 false positives and 2 false negatives on a validation set. Model B makes 6 false positives and 5 false negatives. If a false negative costs 10 units and a false positive costs 1 unit(s), which model has lower expected validation cost and what are the two costs?机器学习中等数值题未尝试面试订阅5061Why Nested Evaluation MattersWhy is it unfair to compare two tuned models on the same validation folds that were also used to pick their hyperparameters?机器学习困难essay未尝试面试订阅5062Why Dependence MattersWhy can standard iid significance arguments be too optimistic when model scores come from overlapping rolling windows?机器学习困难essay未尝试面试订阅5063Metric Choice Can Flip WinnerWhy can a model that wins on AUC still lose on business utility?机器学习困难essay未尝试面试订阅5064Multiple Comparisons TrapWhy should you be skeptical of the best model out of a large tournament even if its validation score is clearly the highest?机器学习困难essay未尝试面试订阅5065Why One Number Is Not EnoughWhy is a single held-out score often not enough evidence to declare one model categorically better?机器学习困难essay未尝试面试订阅5086RL Training Diagnostic 21Why can bootstrapping help value estimates even before an episode terminates?机器学习困难essay未尝试面试订阅5087RL Training Diagnostic 22Why does an RL agent usually need explicit exploration even if its current greedy action already looks good?机器学习困难essay未尝试面试订阅5088Discount Factor IntuitionWhy does increasing the discount factor often make value estimates more sensitive to long-run model misspecification?机器学习困难essay未尝试面试订阅5089RL Training Diagnostic 23Why can off-policy learning become fragile when function approximation, bootstrapping, and distribution shift all interact?机器学习困难essay未尝试面试订阅5090RL In Trading CautionWhy should a quant be careful when mapping a toy MDP intuition directly into live trading?机器学习困难essay未尝试面试订阅5191Bull Call Spread 1You buy a call with strike 100 for premium 4.5 and sell a call with strike 110 for premium 1.5, where 110>100. What are the net debit, the break-even stock price, and the maximum profit at expiry?金融与交易中等数值题未尝试面试订阅5192Bull Call Spread 2You buy a call with strike 95 for premium 6 and sell a call with strike 105 for premium 2.2, where 105>95. What are the net debit, the break-even stock price, and the maximum profit at expiry?金融与交易中等数值题未尝试面试订阅5196Protective Put PnL 1You own the stock at 100 and buy a put with strike 95 for premium 4. What is the strategy's profit at expiry if the stock ends at 92?金融与交易中等数值题未尝试面试订阅5200Protective Put PnL 5You own the stock at 90 and buy a put with strike 85 for premium 3.5. What is the strategy's profit at expiry if the stock ends at 87?金融与交易中等数值题未尝试面试订阅5201Covered Call PnL 1You own the stock at 100 and sell a call with strike 105 for premium 4. What is the strategy's profit at expiry if the stock ends at 112?金融与交易中等数值题未尝试面试订阅5205Covered Call PnL 5You own the stock at 95 and sell a call with strike 100 for premium 3.5. What is the strategy's profit at expiry if the stock ends at 110?金融与交易中等数值题未尝试面试订阅