2024 July07: IIT M AN EXAM QDD2 - 07 Jul 2024

Concepts this paper tests

Auto-generated summary from the source site. Handy as a checklist; the lessons above explain each idea from scratch.

Summary of Core Topics, Concepts, Principles, and Equations

1. Linear Algebra

• Matrices and Vectors: Understanding of matrices, vectors, and their operations.

• Eigenvalues and Eigenvectors: Key concepts for understanding matrix properties.

  • Eigenvalue Equation: $A v = \lambda v$
  • Eigenvector: A non-zero vector $v$ that satisfies the eigenvalue equation.

• Matrix Properties: Symmetric, diagonal, orthogonally diagonalizable matrices.

  • Symmetric Matrix: $A = A^T$
  • Diagonal Matrix: A matrix with non-zero elements only on the main diagonal.
  • Orthogonally Diagonalizable: A matrix that can be diagonalized by an orthogonal matrix.

• Projection Matrix: Properties and calculations.

  • Projection Matrix Formula: $P = \frac{a a^T}{a^T a}$
  • Properties: Symmetric, idempotent, singular.

2. Calculus

• Differentiability: Conditions for a function to be differentiable at a point.

  • Continuity: $\lim_{x \to c} f(x) = f(c)$
  • Differentiability: $\lim_{h \to 0} \frac{f(c+h) - f(c)}{h}$ exists.

• Directional Derivatives: Existence and calculation.

  • Directional Derivative: $D_u f(a, b) = \nabla f(a, b) \cdot u$

• Linear Approximation: Conditions for linear approximation around a point.

  • Linear Approximation Formula: $f(x) \approx f(a) + f'(a)(x-a)$

3. Statistics and Machine Learning

• Least Squares Method: Finding the best-fit line.

  • Least Squares Formula: Minimizing $\sum (y_i - (mx_i + b))^2$

• Classification Models: Misclassification rate and step functions.

  • Misclassification Rate: $\frac{\text{Number of misclassified samples}}{\text{Total number of samples}}$

• Clustering: Identifying inherent groupings in data.

  • Clustering Algorithms: K-means, hierarchical clustering.

4. Specific Equations and Concepts

• Eigenvalue Calculation:

  • For a matrix $A$ and eigenvector $v$, $A v = \lambda v$.

• Projection Matrix:

  • $P = \frac{a a^T}{a^T a}$

• Differentiability Conditions:

  • For a piecewise function, continuity and differentiability at the boundary points.

• Gradient Calculation:

  • $\nabla f = \left( \frac{\partial f}{\partial x}, \frac{\partial f}{\partial y} \right)$
  • Length of Gradient: $\|\nabla f\| = \sqrt{\left( \frac{\partial f}{\partial x} \right)^2 + \left( \frac{\partial f}{\partial y} \right)^2}$

• Least Squares Method:

  • Minimizing the sum of squared residuals.

• Misclassification Rate:

  • Calculating the rate of incorrect predictions.