Mathematics
Orthogonal Polynomials and Their Recurrence Relations
Quick fact
Every family of orthogonal polynomials—like Legendre, Chebyshev, or Hermite—can be generated by a simple three-term recurrence relation, which makes them incredibly efficient to compute and use in numerical algorithms.
Why this is interesting
You know that two lines are perpendicular if they meet at a right angle. But what does it mean for two functions to be perpendicular? And why would that be useful?
Read the full explanation
Understanding Orthogonal Polynomials and Their Recurrence Relations
Think of polynomials as vectors in an infinite-dimensional space. Just as two vectors are orthogonal if their dot product is zero, two polynomials are orthogonal if their inner product—an integral of their product times a weight function—equals zero. This inner product defines a notion of 'angle' and 'length' for functions. For example, the Legendre polynomials are orthogonal on the interval [-1,1] with weight 1: the integral of Pn(x)Pm(x) from -1 to 1 is zero unless n=m. This orthogonality means each polynomial captures a new 'direction' in function space, free from interference from the others. Consequently, any function can be decomposed into a sum of these orthogonal components, much like expressing a vector in terms of perpendicular axes.
A deeper explanation
The magic of orthogonal polynomials is that they can be built step-by-step using a simple three-term recurrence: P{n+1}(x) = (an x + bn) Pn(x) - cn P{n-1}(x). The coefficients an, bn, and cn are determined by the weight function and the normalization. This recurrence is a direct consequence of orthogonality: since xPn(x) is a polynomial of degree n+1, its projection onto the orthogonal basis must involve only P{n+1}, Pn, and P{n-1}. This makes generating them computationally cheap and numerically stable. The same orthogonality also gives rise to Gaussian quadrature, a powerful method for approximating integrals: the optimal nodes are exactly the roots of these polynomials, and the weights are derived from the same theory. Moreover, orthogonal polynomials solve Sturm–Liouville differential equations, which appear throughout physics and engineering, from quantum mechanics (Hermite polynomials for the harmonic oscillator) to fluid dynamics (Chebyshev polynomials for spectral methods). Their recurrence relations are the heart of many algorithms that compute special functions and solve boundary value problems.