Skip to content

📐 Engineering Mathematics – I (100102)

⬅️ Back to Semester-1 | 🏠 Home

💡 Why this subject? Linear algebra + calculus = the math engine behind machine learning, computer graphics, and algorithm analysis you'll see later in CSE.

📌 Unit 1: Linear Algebra – I

  • Elementary Row Operations: swap rows, multiply a row by a constant, add a multiple of one row to another — used to simplify matrices.
  • Gauss-Jordan Method: reduce a matrix [A | I] to [I | A⁻¹] using row operations → gives the inverse directly.
  • Special complex matrices:
  • Hermitian: equal to its own conjugate transpose (A = Aᴴ)
  • Skew-Hermitian: A = -Aᴴ
  • Unitary: AAᴴ = I (like "orthogonal" but for complex numbers)
  • Vector Space & Subspace: a set of vectors closed under addition & scalar multiplication.
  • Linear Independence: vectors that don't "overlap" in direction — none can be written as a combination of the others.
  • Basis & Dimension: a basis is the smallest set of independent vectors that spans the whole space; dimension = number of vectors in that basis.
  • Rank of a Matrix: number of linearly independent rows/columns — tells you how many "real" equations a system actually has.

📝 Example: In a system of 3 equations, if one equation is just 2×(another equation), rank = 2, not 3 — meaning you really only have 2 independent constraints.

🧠 Quick Recall: Rank tells you if a system of equations has a unique solution, infinite solutions, or no solution.

📌 Unit 2: Linear Algebra – II

  • Linear Transformation: a function T: V → W that preserves addition and scalar multiplication — basically, matrices are linear transformations.
  • Kernel (null space): all vectors that map to zero. Range: all possible outputs.
  • Rank-Nullity Theorem: rank(T) + nullity(T) = dimension of domain
  • Eigenvalues & Eigenvectors: for a matrix A, if Av = λv, then v is an eigenvector and λ its eigenvalue — v only gets scaled, not rotated, by A.
  • Diagonalization: rewriting A = PDP⁻¹ where D is diagonal (made of eigenvalues) — makes repeated matrix multiplication (Aⁿ) trivial.
  • Cayley-Hamilton Theorem: every square matrix satisfies its own characteristic equation.

📝 Practical use: Eigenvectors are exactly how Google's PageRank, PCA (data compression), and image compression work — finding the "main direction" of data.

📌 Unit 3: Calculus for Single Variable

  • L'Hospital's Rule: for 0/0 or ∞/∞ limits, differentiate top & bottom separately and try again.
  • Rolle's Theorem: if f(a) = f(b), there's a point in between where f'(x) = 0 (slope is flat somewhere).
  • Mean Value Theorem: there's a point where the instantaneous slope = average slope over the interval.
  • Taylor & Maclaurin Series: approximate any smooth function as an infinite polynomial around a point (Maclaurin = Taylor centered at 0).
  • e^x ≈ 1 + x + x²/2! + x³/3! + ...
  • Beta & Gamma Functions: generalize factorials to non-integers: Γ(n) = (n-1)! for integers.

📝 Example: Calculators compute sin(x) using a truncated Taylor series internally!

📌 Unit 4: Multivariable Calculus (Differentiation)

  • Partial Differentiation: differentiate a multi-variable function w.r.t. one variable, treating others as constants.
  • Jacobian: a matrix of all partial derivatives — used when changing variables (e.g., Cartesian → polar).
  • Maxima/Minima (2 variables): solve fx = 0, fy = 0, then use the second derivative test (Hessian) to classify max/min/saddle.
  • Lagrange Multipliers: find max/min of f(x,y) subject to a constraint g(x,y) = 0 — used in optimization problems (e.g., maximize area given fixed perimeter).

📝 Example: "Maximize box volume given fixed surface area" → classic Lagrange multiplier problem.

📌 Unit 5: Multivariable Calculus (Integration)

  • Double Integral: integrates a function over a 2D region — gives volume under a surface.
  • Triple Integral: integrates over a 3D region — gives total mass/volume.
  • Change of variables: switch to polar (2D), cylindrical, or spherical (3D) coordinates to simplify integrals over circular/spherical regions.

📝 Example: Finding area of a circle is painful in x,y but trivial in polar coordinates: ∫∫ r dr dθ.

📌 Unit 6: Vector Calculus

  • Gradient: points in direction of steepest ascent of a scalar field.
  • Divergence & Curl: measure outward flow and rotation of a vector field (same as in Physics Unit 4!).
  • Line Integral: integrate along a curve/path (e.g., work done by a force along a path).
  • Green's Theorem: converts a line integral around a closed curve into a double integral over the enclosed region.
  • Stokes' Theorem: relates a surface integral of curl to a line integral around its boundary.
  • Gauss-Divergence Theorem: relates a volume integral of divergence to a flux integral over the boundary surface.

🧠 Quick Recall: Green's, Stokes', and Gauss's theorems are all the same idea: "what happens inside = what happens on the boundary."

✅ Quick Revision Table

Topic One-line memory hook
Rank Number of "truly independent" equations
Eigenvectors Directions that only get scaled, not rotated
Diagonalization Makes Aⁿ easy to compute
L'Hospital Use only for 0/0 or ∞/∞
Taylor Series Function ≈ infinite polynomial
Jacobian Matrix of partial derivatives, used in coordinate change
Lagrange Multipliers Optimize with a constraint
Green/Stokes/Gauss Boundary integral = interior integral