kakuritsu

Gaussian Distribution

The normal distribution, often called the bell curve, is the most important probability distribution in statistics. Discovered by Carl Friedrich Gauss, it describes how values cluster around a mean, with the density decreasing symmetrically as you move away from the center.

In nature, countless phenomena follow this pattern: human heights, measurement errors, thermal noise in electronics, and the positions of particles undergoing diffusion. The central limit theorem explains why: when many independent random variables are summed, their normalized sum tends toward a normal distribution, regardless of the original distributions.

P(x) = 1/(σ√2π) · e^{-(x-μ)²/2σ²}

The two parameters μ (mean) and σ (standard deviation) fully describe the distribution. Approximately 68% of values fall within one standard deviation of the mean, 95% within two, and 99.7% within three -- the famous 68-95-99.7 rule.

Random Walks

A random walk is a mathematical formalization of a path consisting of successive random steps. First described by Karl Pearson in 1905, it models everything from the Brownian motion of pollen grains in water to the fluctuations of stock prices on Wall Street.

In its simplest form, a walker on a number line flips a coin at each step: heads moves right, tails moves left. Despite the simplicity, random walks possess remarkable properties. In one and two dimensions, a random walker will eventually return to any previously visited point -- a property called recurrence. In three or more dimensions, this no longer holds.

E[D_n] = √(2n/π)

The expected distance from the origin after n steps grows as the square root of n. This square-root scaling is fundamental to diffusion processes throughout physics, biology, and finance. It explains why a drop of ink spreads slowly in still water: molecules must random-walk their way outward.

Voronoi Tessellation

A Voronoi diagram partitions a plane into regions based on distance to a set of seed points. Each region contains all points closer to its seed than to any other. Named after Georgy Voronoi, these structures appear throughout nature and computation.

Giraffes' coat patterns, the arrangement of cells in biological tissue, the territories of competing organisms, and the coverage areas of cell phone towers all follow Voronoi geometry. The dual of a Voronoi diagram is the Delaunay triangulation, which maximizes the minimum angle of all triangles and is widely used in computational geometry and finite element methods.

V(p_i) = {x : d(x, p_i) ≤ d(x, p_j) ∀ j ≠ i}

When seed points are placed randomly, the resulting Voronoi cells have fascinating statistical properties. The average number of edges per cell converges to 6, a consequence of Euler's formula for planar graphs. The distribution of cell areas follows a gamma distribution.

Monte Carlo Method

The Monte Carlo method uses random sampling to obtain numerical results for problems that might be deterministic in principle. Invented by Stanislaw Ulam and John von Neumann during the Manhattan Project, it was named after the famous casino in Monaco.

The classic demonstration estimates π by throwing random points into a square containing an inscribed circle. The ratio of points landing inside the circle to the total number of points converges to π/4. With enough random samples, any desired precision can be achieved.

π ≈ 4 · (points inside circle / total points)

Modern Monte Carlo methods power applications from financial risk analysis and weather prediction to protein folding simulation and computer graphics. Path tracing in modern film rendering is essentially a Monte Carlo integration over all possible light paths in a scene. The method transforms randomness from a nuisance into a computational tool.