IIT INTEGRATION BEE

Complete Master Notes: Tricks, Results & Standard Integrals

SECTION 1: STANDARD RESULTS FROM YOUR NOTES

Result 1 — Rational Power Integral

INT_0^inf x^p / (x^n + 1) dx = pi / [ n * sin((p+1)pi/n) ]
Condition: 0 < p+1 < n

Special case (p = 0):

INT_0^inf dx / (x^n + 1) = pi / [ n * sin(pi/n) ]

n Result
2 pi/2

3 2pi/(3*sqrt(3))

4 pi/(2*sqrt(2))

6 pi/3

KEY ENGINE: Euler Reflection Formula => GAMMA(s)*GAMMA(1-s) = pi/sin(pi*s)

DERIVATION: Sub x^n = t, gives (1/n)*INT t^(1/n - 1)/(1+t) dt = pi/(n*sin(pi/n))

Result 2 — General Power Integral

INT_0^inf x^p / (x^n + 1) dx = pi / [ n * sin((p+1)pi/n) ]
More general form of Result 1 — covers all rational/irrational p

Result 3 — Gamma Function Power Integral

INT_0^inf x^(m-1) * e^(-a*x^n) dx = GAMMA(m/n) / [ n * a^(m/n) ]

Special Gamma values to memorize:

Value Result
GAMMA(1/2) sqrt(pi)

GAMMA(1) 1

GAMMA(n+1) n! (for integer n)

 Value Result
GAMMA(n+1) = n * GAMMA(n) [recurrence]

GAMMA(n+1/2) (2n)! * sqrt(pi) / (4^n * n!)

EXAMPLE: INT_0^inf e^(-x^5/2025) * x^1.5 dx => m=5/2, n=5, a=1/2025 => 9*sqrt(pi)

Result 4 — Laplace's Method (Peak Integrals)

lim_{n->inf} sqrt(n) * INT_0^inf (e^(x-1)/x^x)^n dx = sqrt(2*pi)

General Laplace principle: For I_n = INT e^(n*g(x)) dx peaked at x_0:

I_n ~ e^(n*g(x_0)) * sqrt( 2*pi / (n*|g''(x_0)|) )

Recipe:
• Write integrand as e^(n*g(x))
• Find maximum x_0: solve g'(x_0) = 0
• Compute g''(x_0)
• Gaussian integral gives the sqrt(2*pi/n) factor
CONNECTION: Same sqrt(2*pi) as Stirling's formula n! ~ sqrt(2*pi*n)*(n/e)^n

Result 5 — Dilogarithm (Li_2)

INT_0^(1/2) log(1-x)/x dx = (ln^2(2))/2 - pi^2/12

Key identity used (Reflection Formula):

Li_2(x) + Li_2(1-x) = pi^2/6 - ln(x)*ln(1-x)

Set x = 1/2:
Li_2(1/2) = pi^2/12 - (ln^2(2))/2
Li_2(x) = SUM_{n=1}^inf x^n/n^2 (dilogarithm / Spence's function)
DERIVATION: d/dx[Li_2(x)+Li_2(1-x)] = -ln(1-x)/x + ln(x)/(1-x) = d/dx[-ln(x)ln(1-x)]
SECTION 2: HOLY TRINITY — INVERSE TRIG INTEGRALS
Integral Result
INT dx / sqrt(a^2 - x^2) arcsin(x/a) + C

INT dx / (a^2 + x^2) (1/a) arctan(x/a) + C

INT dx / [x*sqrt(x^2 - a^2)] (1/a) sec^(-1)(x/a) + C

NOTE: sec^(-1)(t/a) = arctan(sqrt(t^2-a^2)/a) — both forms are identical!

SECTION 3: KNOWN SERIES RESULTS

Series Value Name
SUM_{n=1}^inf 1/n^2 pi^2/6 Basel Problem (zeta(2))

SUM_{n=1}^inf 1/n^4 pi^4/90 zeta(4)

SUM_{n=1}^inf (-1)^(n-1)/n^2 pi^2/12 Dirichlet eta(2)

SUM_{n=1}^inf (-1)^(n-1)/n ln(2) Alternating harmonic

SUM_{n=1}^inf 1/n^2/2^n pi^2/12 - ln^2(2)/2 Li_2(1/2)

SUM_{n=0}^inf (-1)^n/(2n+1) pi/4 Leibniz formula

eta(s) = (1 - 2^(1-s)) * zeta(s) [Dirichlet eta vs Riemann zeta relation]

INT_0^inf x^3/(e^x-1) dx = 6*zeta(4) = pi^4/15 [Bose-Einstein / blackbody]
INT_0^inf x^3/(e^x+1) dx = 6*(7/8)*zeta(4) = 7*pi^4/120 [Fermi-Dirac]

SECTION 4: EXPANSION & SERIES TRICKS

4.1 Geometric Series Expansion (Most Used!)

1/(1+x) = SUM_{n=0}^inf (-x)^n = 1 - x + x^2 - x^3 + ... |x|<1
1/(1-x) = SUM_{n=0}^inf x^n = 1 + x + x^2 + x^3 + ... |x|<1

Classic application:
log(1-x)/x = -SUM_{n=1}^inf x^(n-1)/n
Then INT_0^1 log(1-x)/x dx = -SUM 1/n^2 = -pi^2/6
4.2 Exponential & Trig Expansions
e^x = SUM_{n=0}^inf x^n/n!
sin(x) = SUM_{n=0}^inf (-1)^n x^(2n+1)/(2n+1)!
cos(x) = SUM_{n=0}^inf (-1)^n x^(2n)/(2n)!
ln(1+x) = SUM_{n=1}^inf (-1)^(n-1) x^n/n

4.3 Feynman's Trick (Differentiation Under Integral Sign)

d/da INT f(x,a) dx = INT (df/da) dx

Example: INT_0^1 (x^a - 1)/ln(x) dx = ln(a+1)

Strategy: introduce parameter a, differentiate, integrate in x, then integrate in a
Works best when a appears in an exponent or inside a log

4.4 King's Property

INT_a^b f(x) dx = INT_a^b f(a+b-x) dx

Add I + I using substitution x -> a+b-x, often the sum simplifies dramatically
EXAMPLE: INT_0^(pi/2) ln(sin x) dx = INT_0^(pi/2) ln(cos x) dx = -pi*ln(2)/2

4.5 Riemann Sum to Integral

lim_{n->inf} (1/n) SUM_{k=1}^n f(k/n) = INT_0^1 f(x) dx

Always factor out 1/n first, then identify f(k/n)

Can also switch order of summation/integration for double integrals

4.6 Layer Cake / Cavalieri's Formula

INT f(x) dx = INT_0^inf mu{x: f(x) >= t} dt

floor(f(x)) = SUM_{k=1}^inf 1[f(x) >= k]

Used in: floor integral problems, measure-theoretic integrals
SECTION 5: SUBSTITUTION TRICKS

5.1 The t^4+1 Split Trick (VERY IMPORTANT!)

x^2/(x^4+1) = (1/2)*(x^2+1)/(x^4+1) + (1/2)*(x^2-1)/(x^4+1)

Then divide top and bottom by x^2:

(1 + 1/x^2) / (x^2 + 1/x^2) => u = x - 1/x, u^2 = x^2 - 2 + 1/x^2
(1 - 1/x^2) / (x^2 + 1/x^2) => v = x + 1/x, v^2 = x^2 + 2 + 1/x^2

x^2 + 1/x^2 = (x - 1/x)^2 + 2 OR (x + 1/x)^2 - 2

APPLICATION: INT sqrt(tan x) dx => sub t = sqrt(tan x) => 2*INT t^2/(t^4+1) dt

5.2 Weierstrass / Half-Angle Substitution

t = tan(x/2) => sin x = 2t/(1+t^2), cos x = (1-t^2)/(1+t^2), dx =
2dt/(1+t^2)
Best for rational functions of sin x and cos x

5.3 Euler Substitution

For integrals with sqrt(ax^2 + bx + c):
• Type 1: a > 0: sqrt(ax^2+bx+c) = t - sqrt(a)*x
• Type 2: c > 0: sqrt(ax^2+bx+c) = tx + sqrt(c)
• Type 3: real roots r,s: sqrt(a(x-r)(x-s)) = t(x-r)

5.4 Beta Function Substitution

B(m,n) = INT_0^1 x^(m-1)(1-x)^(n-1) dx = GAMMA(m)*GAMMA(n)/GAMMA(m+n)

INT_0^inf x^(s-1)/(1+x) dx = pi/sin(pi*s) [0 < s < 1]

This is the key step: sub x^n = t in Result 1 to get this form

5.5 Chebyshev Substitution

x = 2*cos(theta) => x^2 - 2 = 2*cos(2*theta)
After n squarings: nested ((...((x^2-2)^2-2)...)^2) = 2*cos(2^n * theta)
Used for: nested squaring integrals, Chebyshev polynomials

5.6 Triple Angle Trick

(3x - x^3)/(1 - 3x^2) = tan(3*arctan(x))
So arctan((3x-x^3)/(1-3x^2)) = 3*arctan(x) ONLY when |arctan(x)| < pi/6
CRITICAL: Must split at x = 1/sqrt(3) where formula breaks! Add -pi correction term.

5.7 Residue/Vieta Trick for f(x) Integrals

INT_{-inf}^{inf} f(x)^2/(1+f(x)^2) dx = pi * SUM_j a_j
For f(x) = SUM a_j/(x-c_j):
• Set f(x) = u, find polynomial equation for x
• Use Vieta's formulas for sum of roots = h(u)
• Differentiate: SUM 1/f'(x_j) = d/du [sum of roots]
• The u^2 cancels, leaving pi * SUM(a_j)

SECTION 6: CONTOUR INTEGRATION (Complex Analysis)

6.1 Basic Setup

• Replace cos(ax) with Re(e^(iax)), sin(ax) with Im(e^(iax))
• Close contour in UPPER half plane when a > 0 (e^(iaz) decays there)
• Close in LOWER half plane when a < 0

INT_{-inf}^{inf} f(x) dx = 2*pi*i * SUM [Res at upper half plane poles]

6.2 Residue at Simple Pole

Res_{z=a} f(z)/g(z) = f(a)/g'(a)

6.3 Sophie Germain Identity

x^4 + 4a^4 = (x^2 + 2a^2 + 2ax)(x^2 + 2a^2 - 2ax)
Special case a=1: x^4+4 = (x^2+2x+2)(x^2-2x+2)
Poles of 1/(z^4+4): at z = 1+i, -1+i (upper), 1-i, -1-i (lower)
SECTION 7: INTEGRATION BY PARTS & REDUCTION

7.1 LIATE Rule

Choose u in this priority order:
Letter Type Example
L Logarithmic ln x, log x

I Inverse trig arctan x, arcsin x

A Algebraic x^n, polynomials

T Trigonometric sin x, cos x

E Exponential e^x, a^x

7.2 Key IBP Results

INT arctan(x) dx = x*arctan(x) - (1/2)*ln(1+x^2) + C
INT x^n * e^x dx = e^x * SUM_{k=0}^n (-1)^k * n!/(n-k)! * x^(n-k) + C
For integer n: IBP terminates after n+1 steps
For non-integer n: result involves incomplete Gamma or erfi

7.3 Reduction Formulas

INT sin^n(x) dx = -sin^(n-1)(x)cos(x)/n + (n-1)/n * INT sin^(n-2)(x) dx
INT cos^n(x) dx = cos^(n-1)(x)sin(x)/n + (n-1)/n * INT cos^(n-2)(x) dx
INT sec^n(x) dx = sec^(n-2)(x)tan(x)/(n-1) + (n-2)/(n-1) * INT sec^(n-2)(x)
dx

SECTION 8: PARTIAL FRACTIONS

8.1 Types of Decomposition

Factor in denominator Partial fraction terms
(x-a) A/(x-a)

(x-a)^2 A/(x-a) + B/(x-a)^2

(x^2+bx+c) irreducible (Ax+B)/(x^2+bx+c)

(x^2+bx+c)^2 (Ax+B)/(x^2+bx+c) + (Cx+D)/(x^2+bx+c)^2

8.2 Irreducible Quadratic — Always Do This

Complete square: x^2 + bx + c = (x + b/2)^2 + (c - b^2/4)
Split numerator: write px + q = (p/2)*(2x+b) + (q - pb/2)
First part: numerator = derivative of denominator => gives ln
Second part: constant numerator => gives arctan

SECTION 9: MIN/MAX & FLOOR INTEGRALS

9.1 Min/Max Strategy

• Find crossover points: solve f(x) = g(x)
• Build sign table on each subinterval
• Split integral at crossover points
• Watch for wrap-around corrections (like -pi in arctan triple angle)

EXAMPLE: INT_0^pi min(cos x, cos 2x) dx = -3*sqrt(3)/4

Crossover at x = 2*pi/3 (where cos x = cos 2x = -1/2)

9.2 Floor Function Strategy — Layer Cake

INT floor(f(x)) dx = SUM_{k=1}^inf mu{ x : f(x) >= k }

Near a simple pole where f ~ C/|x-a|: contribution = C*pi^2/3

EXAMPLE: floor integral => SUM 1/n^2 = pi^2/6 appears via Basel series

9.3 Number Theory in Floor Integrals

SUM_{k=1}^inf floor(x/k) = SUM_{m<=x} d(m) [d(m) = number of divisors]

d(m) is ODD iff m is a perfect square!

So (-1)^{SUM floor(x/k)} = (-1)^{floor(sqrt(x))}
This reduced the scary exponent to just floor(sqrt(x)) in our problem
SECTION 10: SPECIAL FUNCTIONS CHEAT SHEET
Function Definition Key Property
GAMMA(s) INT_0^inf t^(s-1) e^(-t) dt GAMMA(n+1) = n!

B(m,n) INT_0^1 t^(m-1)(1-t)^(n-1) dt =

GAMMA(m)GAMMA(n)/GAMMA(m+
n)

Li_2(x) SUM x^n/n^2 = -INT_0^x ln(1-t)/t dt Li_2(1/2) = pi^2/12 - ln^2(2)/2

erf(x) (2/sqrt(pi)) INT_0^x e^(-t^2) dt erf(inf) = 1

erfi(x) (2/sqrt(pi)) INT_0^x e^(t^2) dt Appears in INT sqrt(x)*e^x dx

Ei(x) -P.V. INT_{-x}^inf e^(-t)/t dt = INT e^x/x dx

gamma(s,x) INT_0^x t^(s-1) e^(-t) dt Lower incomplete gamma

GAMMA(s,x) INT_x^inf t^(s-1) e^(-t) dt Upper incomplete gamma

SECTION 11: TRIG IDENTITIES FOR INTEGRATION

11.1 Power Reduction

sin^2(x) = (1 - cos(2x))/2
cos^2(x) = (1 + cos(2x))/2
sin^3(x) = (3*sin(x) - sin(3x))/4
cos^4(x) = (3 + 4*cos(2x) + cos(4x))/8

11.2 Product to Sum

sin(A)sin(B) = [cos(A-B) - cos(A+B)]/2
cos(A)cos(B) = [cos(A-B) + cos(A+B)]/2
sin(A)cos(B) = [sin(A+B) + sin(A-B)]/2

11.3 Key Substitution Identities

sqrt((1-sin x)/(1+sin x)) = tan(pi/4 - x/2)
sqrt((1-cos x)/(1+cos x)) = tan(x/2)
(3x - x^3)/(1-3x^2) = tan(3*arctan x) [triple angle]
(2x)/(1-x^2) = tan(2*arctan x) [double angle]

11.4 Wallis Integral

INT_0^(pi/2) sin^n(x) dx = INT_0^(pi/2) cos^n(x) dx
= (pi/2) * [(n-1)!!/(n!!)] [for even n]
 = [(n-1)!!/(n!!)] [for odd n]
n!! = double factorial: n*(n-2)*(n-4)*...*2 or *1

SECTION 12: IMPORTANT DEFINITE INTEGRAL RESULTS

Integral Result
INT_0^(pi/2) ln(sin x) dx -pi*ln(2)/2

INT_0^(pi/2) ln(cos x) dx -pi*ln(2)/2

INT_0^inf sin(x)/x dx pi/2 [Dirichlet integral]

INT_0^inf e^(-x^2) dx sqrt(pi)/2 [Gaussian]

INT_-inf^inf e^(-x^2) dx sqrt(pi)

INT_0^inf x^3/(e^x-1) dx pi^4/15 [Bose-Einstein]

INT_0^inf x^3/(e^x+1) dx 7*pi^4/120 [Fermi-Dirac]

INT_0^1 log(1-x)/x dx -pi^2/6

INT_0^(1/2) log(1-x)/x dx ln^2(2)/2 - pi^2/12

INT_0^inf dx/(1+x^2) pi/2

INT_0^inf x^(s-1)/(1+x) dx pi/sin(pi*s)

INT_0^pi min(cos x, cos 2x) dx -3*sqrt(3)/4

INT_0^(sqrt(3)) arctan((3x-x^3)/(1-3x^2)) dx pi/sqrt(3) - 3*ln(2)

SECTION 13: QUICK COMPETITION STRATEGIES

When you see...

Integrand form Strategy
x^p/(x^n+1) Beta function / Euler reflection => pi/[n*sin((p+1)pi/n)]

e^(-ax^n) * x^m Gamma function => GAMMA((m+1)/n) / [n * a^((m+1)/n)]

f(x)^n as n->inf Laplace's method: expand g=ln(f) around max

SUM 1/n with (-1)^n Dirichlet eta = (1-2^(1-s))*zeta(s)

1/(t^4+1) type Split trick: x^2+1 and x^2-1 in numerator

sqrt(tan x) Sub t=sqrt(tan x) => 2t^2/(t^4+1) => split trick

arctan(rational) Check if argument is tan(n*theta): triple/double angle

Rational + sqrt quadratic Euler substitution or complete the square

Integrand form Strategy
Floor/ceiling Layer cake: SUM of measures

f(x)=SUM a_j/(x-c_j) Vieta trick => integral = pi*SUM(a_j)

cos(ax)/(poly) on real line Contour: replace with e^(iaz), residues

Nested min/max Work inside out, build sign table

x->a+b-x symmetric King's property: add I+I

log inside integral [0,1] Expand log as series, integrate term by term

Riemann sum pattern Factor 1/n, identify f(k/n), convert to integral

SECTION 14: COMMON MISTAKES TO AVOID

• arctan triple angle: ALWAYS split at x=1/sqrt(3), add -pi correction in (1/sqrt(3), sqrt(3))
• Contour: check which half-plane e^(iaz) decays in (upper if a>0)
• Partial fractions: for repeated roots, include ALL powers up to the multiplicity
• Floor integrals: measure near pole is 2C/n^2, not C/n^2 (factor of 2 for both sides)
• Laplace method: expand to second order, not first!
• King's property: only works when limits are symmetric or satisfy a+b = const
• x^m * e^x indefinite: only closes for non-negative integer m; otherwise use special functions
• Series swap and integral: always verify uniform convergence before swapping

Good luck at the IIT Integration Bee! All results compiled from your session.