Generating Unique Random Numbers

Pseudorandom numbers

Pseudorandom numbers are values generated by algorithms. These values are determined by a seed, or can be produced from user input e.g. mouse movement, time between keystrokes.

Hence, using the same seed in the same algorithm will generate the same "random" numbers.

Uses of Random Numbers

• Cryptography: generating an encryption key requires randomness for high entropy (i.e. high unpredictability)

• Games: procedural-generated utilise random numbers to create a unique feel to the game, such as terrain generation (Perlin noise)

• Simulations: random numbers are useful for simulating real world models and as training data for AI

• Experiments: experiments require values that are unbiased, and thus random numbers can be used

Definitions

• Perlin noise: Perlin noise is a form of gradient noise used to differentiate the "machine-like" look of computer graphics.

• Seed: A seed is a number or vector that instantiates the way in which the algorithm generates a sequence of numbers.