Level 1 — RecognitionAdvanced Topics (Elite Level)

Advanced Topics (Elite Level)

20 minutes40 marksprintable — key stays hidden on paper

Time limit: 20 minutes Total marks: 40 Instructions: Answer all questions. For True/False and matching items, a brief justification is required to earn full marks. Use ...... notation for mathematics.


Section A — Multiple Choice (2 marks each)

Q1. The Cauchy–Riemann equations for f(z)=u(x,y)+iv(x,y)f(z)=u(x,y)+iv(x,y) are:

  • (a) ux=vy, uy=vxu_x=v_y,\ u_y=v_x
  • (b) ux=vy, uy=vxu_x=v_y,\ u_y=-v_x
  • (c) ux=vy, uy=vxu_x=-v_y,\ u_y=v_x
  • (d) ux=uy, vx=vyu_x=u_y,\ v_x=v_y

Q2. By Cauchy's integral theorem, Cf(z)dz=0\oint_C f(z)\,dz=0 provided:

  • (a) ff is bounded on CC
  • (b) ff is analytic on and inside the closed contour CC
  • (c) CC is a straight line
  • (d) ff has exactly one pole inside CC

Q3. The residue of f(z)=1z2+1f(z)=\dfrac{1}{z^2+1} at z=iz=i is:

  • (a) 12i\tfrac{1}{2i}
  • (b) 12i-\tfrac{1}{2i}
  • (c) ii
  • (d) 12\tfrac{1}{2}

Q4. In the Einstein summation convention, aibia_i b^i denotes:

  • (a) a single product term
  • (b) the sum iaibi\sum_i a_i b^i over the repeated index
  • (c) a rank-2 tensor
  • (d) an undefined expression

Q5. The Euler–Lagrange equation for J[y]=abF(x,y,y)dxJ[y]=\int_a^b F(x,y,y')\,dx is:

  • (a) Fy=0\dfrac{\partial F}{\partial y}=0
  • (b) FyddxFy=0\dfrac{\partial F}{\partial y}-\dfrac{d}{dx}\dfrac{\partial F}{\partial y'}=0
  • (c) ddxFy=0\dfrac{d}{dx}\dfrac{\partial F}{\partial y}=0
  • (d) Fy=0\dfrac{\partial F}{\partial y'}=0

Q6. The metric tensor gijg_{ij} is used to:

  • (a) differentiate scalars
  • (b) lower an index, converting a contravariant component to covariant
  • (c) count dimensions
  • (d) diagonalize any matrix

Q7. A function ff is convex on a convex set if for all x,yx,y and λ[0,1]\lambda\in[0,1]:

  • (a) f(λx+(1λ)y)λf(x)+(1λ)f(y)f(\lambda x+(1-\lambda)y)\ge \lambda f(x)+(1-\lambda)f(y)
  • (b) f(λx+(1λ)y)λf(x)+(1λ)f(y)f(\lambda x+(1-\lambda)y)\le \lambda f(x)+(1-\lambda)f(y)
  • (c) f(x)<0f''(x)<0 everywhere
  • (d) ff is linear

Q8. The stationary curve minimizing descent time in the brachistochrone problem is:

  • (a) a straight line
  • (b) a parabola
  • (c) a cycloid
  • (d) a circular arc

Q9. For a Markov chain with transition matrix PP, the steady-state (stationary) distribution π\pi satisfies:

  • (a) πP=π\pi P=\pi
  • (b) Pπ=0P\pi=0
  • (c) π=P\pi=P
  • (d) π2=π\pi^2=\pi

Q10. The KKT conditions for a constrained minimization include, for inequality constraint g(x)0g(x)\le0 with multiplier μ\mu:

  • (a) μ<0\mu<0
  • (b) complementary slackness μg(x)=0\mu\, g(x)=0 and μ0\mu\ge0
  • (c) g(x)=0g(x)=0 always
  • (d) f=0\nabla f=0 only

Section B — True/False WITH Justification (2 marks each: 1 verdict + 1 reason)

Q11. "Every function differentiable at a point in the complex sense is analytic there." — True or False? Justify.

Q12. "Uniform continuity of ff on a set implies pointwise continuity." — True or False? Justify.

Q13. "A function sequence fnff_n\to f pointwise necessarily converges uniformly." — True or False? Justify.

Q14. "The Cauchy–Tukey FFT reduces the DFT of NN points (with NN a power of 2) from O(N2)O(N^2) to O(NlogN)O(N\log N) operations." — True or False? Justify.

Q15. "A pole of order 2 has a principal part in its Laurent expansion containing exactly one negative-power term." — True or False? Justify.


Section C — Matching (5 marks total, 1 each)

Q16. Match each concept (A–E) to its correct description (1–5).

Concept Description
A. Residue theorem 1. Symmetric object measuring lengths/angles in a manifold
B. Christoffel symbols 2. Cfdz=2πiRes\oint_C f\,dz = 2\pi i\sum \text{Res} inside CC
C. Metric tensor 3. Iterative first-order minimization xk+1=xkηfx_{k+1}=x_k-\eta\nabla f
D. Gradient descent 4. Encode how basis vectors change; connection coefficients
E. Lebesgue measure 5. Assigns "size" to subsets of Rn\mathbb{R}^n, generalizing length
Answer keyMark scheme & solutions

Section A

Q1. (b). CR equations require ux=vyu_x=v_y and uy=vxu_y=-v_x; these come from equating the two directional limits of f(z)f'(z). (2)

Q2. (b). Cauchy's theorem needs analyticity on and inside a closed contour in a simply-connected region. (2)

Q3. (a). Simple pole: Resz=i1(zi)(z+i)=12i\text{Res}_{z=i}\frac{1}{(z-i)(z+i)}=\frac{1}{2i}. (2)

Q4. (b). A repeated upper/lower index pair implies summation. (2)

Q5. (b). Standard Euler–Lagrange stationarity condition. (2)

Q6. (b). vi=gijvjv_i=g_{ij}v^j lowers an index. (2)

Q7. (b). Convexity means the graph lies below chords: Jensen inequality with \le. (2)

Q8. (c). The brachistochrone extremal is a cycloid. (2)

Q9. (a). Left eigenvector with eigenvalue 1: πP=π\pi P=\pi (row vector convention). (2)

Q10. (b). KKT: stationarity, primal/dual feasibility (μ0\mu\ge0), and complementary slackness μg=0\mu g=0. (2)

Section B

Q11. False. Analyticity requires complex differentiability in an open neighborhood, not merely at a single point. E.g. f(z)=z2f(z)=|z|^2 is differentiable only at z=0z=0 but nowhere analytic. (1 verdict + 1 reason)

Q12. True. Uniform continuity uses a δ\delta independent of the point, which in particular works at each fixed point; hence it implies (indeed is stronger than) pointwise continuity. (2)

Q13. False. Pointwise ≠ uniform. Counterexample: fn(x)=xnf_n(x)=x^n on [0,1][0,1] converges pointwise to a discontinuous limit, but not uniformly. (2)

Q14. True. Cooley–Tukey recursively splits the DFT (even/odd indices), giving the recurrence T(N)=2T(N/2)+O(N)T(N)=2T(N/2)+O(N), i.e. O(NlogN)O(N\log N). (2)

Q15. False. An order-2 pole has principal part a2(zz0)2+a1(zz0)\frac{a_{-2}}{(z-z_0)^2}+\frac{a_{-1}}{(z-z_0)} with a20a_{-2}\ne0 — generally two negative-power terms. (2)

Section C

Q16. A–2, B–4, C–1, D–3, E–5. (1 mark each correct pairing.) (5)

[
  {"claim":"Residue of 1/(z^2+1) at z=i is 1/(2i)","code":"z=symbols('z'); r=residue(1/(z**2+1), z, I); result = simplify(r - 1/(2*I))==0"},
  {"claim":"Metric lowers index consistency: Res at z=-i is -1/(2i) confirming symmetry","code":"z=symbols('z'); r=residue(1/(z**2+1), z, -I); result = simplify(r + 1/(2*I))==0"},
  {"claim":"Convexity: for f(x)=x^2 midpoint value <= average of endpoints","code":"x,y=symbols('x y'); lhs=((x+y)/2)**2; rhs=(x**2+y**2)/2; result = simplify(rhs-lhs) == ((x-y)**2)/4 and (((x-y)**2)/4).subs({x:1,y:0})>=0"},
  {"claim":"x^n at x=1 stays 1 while limit for x<1 is 0 (nonuniform)","code":"n=symbols('n', positive=True); result = (limit(Rational(1,2)**n, n, oo)==0) and (1**n==1)"}
]