第 84 / 87 页
非代码面试题
显示 20 / 1721 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
5892Posterior from a Generative Gaussian ModelA generative classifier models one feature as Gaussian within each class with equal variance: x|Y=0 ~ N(0,1), x|Y=1 ~ N(2,1), and class prior P(Y=1)=0.5. Using Bayes' rule to convert this generative description into the discriminative posterior, compute P(Y=1|x=1.5).机器学习中等数值题未尝试面试订阅5893Deriving the Even-Money Kelly FractionYou repeatedly bet a fraction f of your current wealth on an even-money wager that wins with probability p>\tfrac12 (you gain the staked amount on a win, lose it on a loss). By maximizing the expected logarithm of your wealth multiplier over one round, derive the growth-optimal fraction f *.概率简单derivation未尝试免费5894Kelly Fraction at General Net OddsA favorable bet pays net odds b to 1: staking an amount, you gain b times the stake with probability p and lose the stake with probability 1-p. Betting a fraction f of wealth each round, derive the growth-optimal fraction f * in terms of b and p.概率简单derivation未尝试免费5895Maximum Growth Rate of a Kelly BettorAn even-money coin wins with probability p=0.6. You bet the growth-optimal (Kelly) fraction every round. Compute the resulting maximum expected log-growth rate per round, and express it in closed form in terms of p.概率中等数值题未尝试免费5903Capping the Single-Bet DrawdownYou bet a fraction f of wealth on an even-money coin with win probability p=0.65, but a risk rule forbids any single losing bet from cutting your wealth by more than 20\%. What fraction should you bet, and for which win probabilities p does this drawdown rule actually constrain you below the Kelly fraction?概率简单数值题未尝试免费5904Kelly Exceeds Full InvestmentA favorable bet has limited downside: staking a fraction f of wealth, you gain the full amount f with probability p=0.7 but lose only half the stake, 0.5f, with probability 0.3. (a) Find the growth-optimal fraction f *. (b) If you cannot borrow (so f\le 1, i.e. you can stake at most all your wealth), what fraction do you actually bet?概率简单数值题未尝试免费5907Kelly with a Proportional Trading CostOn an even-money coin with win probability p, each round you pay a proportional cost c on the amount staked, regardless of the outcome. So staking fraction f, a win multiplies wealth by 1+f(1-c) and a loss by 1-f(1+c). Derive the growth-optimal fraction f * in terms of p and c, evaluate it for p=0.6,\ c=0.05, and find the cost level at which the optimal stake drops to zero.概率困难数值题未尝试面试订阅5916Most You Would Pay for a Perfect TestA product launch pays +30 if the market is receptive and -12 if it is not; receptivity has prior probability 3 10 . You may instead shelve the product for 0. A consultant offers a perfectly accurate test that reveals the true market state before you decide. What is the most you should be willing to pay for this test?概率中等derivation未尝试免费5917Free Peek Before Calling the Bigger BoxTwo boxes each independently contain an amount drawn uniformly from \ 1,2,3,4\ . You must guess which box holds the strictly larger amount; a correct guess pays 1 and a tie or wrong guess pays 0. Before guessing you may take a free peek at the contents of one box (your choice of which). By how much does this peek increase your probability of a correct guess compared with guessing blind?概率简单derivation未尝试免费5918Defective-Batch Inspection With an Imperfect DetectorA batch is defective with prior probability \frac14. Accepting a good batch pays +20; accepting a defective batch pays -40; rejecting pays 0. Before deciding you may run a detector that flags 'defective.' It flags a truly defective batch with probability 9 10 and a good batch with probability \frac15 (false positive). What is the value of running the detector (the increase in expected payoff from using it optimally)?概率困难derivation未尝试面试订阅5919One Free Draw Before Betting on the Majority ColorAn urn is type-R with probability \frac35 (then it is 80\% red balls) or type-B with probability \frac25 (then it is 80\% blue balls). You will bet on the urn's majority color: a correct bet pays 1, a wrong bet pays 0. You may first draw one ball (with replacement) and observe its color for free. By how much does observing this single draw raise your expected payoff over betting with no draw?概率中等derivation未尝试免费5920Clairvoyance Across Three StatesThe state is High, Mid, or Low with probabilities \frac12,\frac13,\frac16. You pick action Long or Flat once. Long pays 12,\ -3,\ -9 in High, Mid, Low respectively; Flat pays 0 in every state. A clairvoyant will tell you the exact state before you choose. What is the difference between your expected payoff acting on the clairvoyant's report and your expected payoff using the single best action chosen in advance?概率中等derivation未尝试免费5921Is the Analyst's Report Worth Its FeeAn investment pays +14 if a deal closes and -10 if it falls through; closing has prior probability \frac12. You may invest or pass (pass pays 0). For a fee of 2 you may buy an analyst report that correctly predicts the outcome with probability 7 10 , after which you decide. Should you buy the report, and what is its value net of the no-report optimum?概率中等derivation未尝试免费5981Total Stake Until the First SixYou roll a fair die repeatedly until the first time a 6 appears; let N be the number of rolls (including the winning roll). On each roll you independently collect a payoff X i with E[X i]=1.5, where the X i are i.i.d. and independent of the roll values. Compute E\! [\sum i=1 N X i ].概率简单derivation未尝试免费5984Expected Inspection Cost Until the First DefectA quality line inspects items one at a time; each item is defective independently with probability 0.05. Inspection stops at the first defective item. Each inspection (defective or not) costs an i.i.d. amount C i with E[C i]=\8, independent of the defect outcomes. Let N be the number of items inspected. Find the expected total inspection cost E\! [\sum i=1 N C i ].概率简单数值题未尝试免费5985Expected Total Slippage With Negative DriftA market-making desk incurs i.i.d. per-trade adverse-selection costs X 1,X 2,\dots with E[X i]=-0.4 (a net loss per trade). The number of trades in a session, N, is independent of the costs and is Poisson with mean 15. Compute the expected cumulative cost E\! [\sum i=1 N X i ].概率简单derivation未尝试免费5987When the Stopping Rule Looks at the Last DrawDraw i.i.d. values X 1,X 2,\dots uniform on \ 1,2,3\ (so E[X i]=2). Define N as follows: keep drawing and stop the first time you draw a 3; let N be the number of draws. Let S N=\sum i=1 N X i. A candidate computes E[N]E[X 1]=3 2=6 and claims E[S N]=6. Compute the correct value of E[S N] and explain in one sentence why E[N]E[X 1] is the wrong formula here.概率困难essay未尝试面试订阅5989Variance of a Count WindowTrades hit a tape as a Poisson process with rate 6 per hour. Let N be the number of trades in a fixed 20-minute window. What is Var (N)?概率简单数值题未尝试免费5990Expected Time to the Third ArrivalOrders arrive at a matching engine as a Poisson process with rate 4 per minute. What is the expected time, in seconds, until the 3rd order arrives (measured from time 0)?概率简单数值题未尝试免费5991Quiet Window on a Combined FeedTwo independent exchanges send quotes to your gateway. Exchange A is a Poisson process with rate 3 per minute and exchange B is an independent Poisson process with rate 5 per minute. Treating the combined stream as one process, what is the probability that no quote arrives during a 30-second window? Give a decimal to three places.概率中等数值题未尝试免费