Mathematics

For data science, 4 topics are important:

1. Linear algebra - 35%
2. Probability and statistics - 25%
3. Calculus - 15%
4. Optimization and formulation of machine-learning algorithms - 25%

Linear Algebra

• Vectors, Scalars, Matrix and Tensor
• Addition, subtraction and product of matrices
• Transpose, rank and determinant and inverse of a matrix
• Dot product of 2 vectors, eigen vectors

Calculus

• Differentiation, Gradient maxima and minima of a function
• Convex and non-convex function, Multivariate function

Probability

• Unions, Intersection, and Conditional probability
• Bayes rule, Probability mass and density function
• Expectation variance and co-variance of a random variable
• Skewness, Kurtosis, Multivariate normal distribution
• Bernoulli, Binomial, Poisson distribution
• Maximum likelihood estimate, Hypothesis testing and p-value

Optimization and formulation

• Gradient and stochastic gradient descent functions
• Constrained optimization problem