S2: 4xy
S7: 6
S12: 0
Q4: 1/2
S22: (2, 3)
S28: Target/Mean Encoding
S29: TimeSeriesSplit
S38: 15
S43: 2/5
S59: Mode > Median > Mean

ML - 2: The data has a Gaussian distribution
ML - 7: Updating prior beliefs with observed data using Bayes' theorem
ML - 12: The probability distribution over actions given states
ML - 17: Internal covariate shift
ML - 23: Boosting reduces bias, bagging reduces variance
ML - 24: Binary Cross-Entropy
S48: 30/84
S60: 150
S53: 2/3
S68: Prior × Likelihood


Amazon Machine Learning Summer School:

Exam Date: 3rd August 2025
Exam Duration: 60 mins

10:30 AM

Test format:
Section 1 - MCQ section - 20 questions
Section 2 - DSA type coding section - 2 questions

Answers available 👇👇👇
Telegram channel:- @allcoding1

M1: Encode sequential order

T1: Regularization

M16: Likelihood × Prior

ML2: SGD can escape local minima due to its noisy updates

ML3: Recursive Feature Elimination (RFE)

ML4: Gini Index

M4: It overfits the training data

M7: 0

M2: Strong negative linear relationship

S13: No real solution
S17: Local minimum
S23: Converges by Limit Comparison with 1/n²
S30: Collect recent user data and evaluate model drift
S31: Data leakage inflated model performance
S39: 6/216

Amazon ML School MCQ Answers 1:15 PM

Kill the enemy
C++
Amazon 1.15 PM


#include <vector>
#include <algorithm>

int solve(std::vector<int> &A, int B) {
long long m1 = 0, m2 = 0;

for (int val : A) {
if (val > m1) {
m2 = m1;
m1 = val;
} else if (val > m2) {
m2 = val;
}
}

long long b = B;
long long s = m1 + m2;

if (s == 0) {
return b > 0 ? -1 : 0;
}

long long k = b / s;
int ans = k * 2;
long long rem = b % s;

if (rem == 0) {
return ans;
} else if (rem <= m1) {
return ans + 1;
} else {
return ans + 2;
}
}

Company: EY ( Ernst & Young)

Role: Tech Analyst

CTC: 12-15 lakhs

Batch: 2025

Apply:
https://careers.ey.com/ey/job/New-Delhi-Analyst-Tech-Consulting-GOV-CNS-TC-Technology-Strategy-&-Transformation-New-Delhi-Nati-110037/1232720901/