quantized

Quantized is an adjective used to describe a quantity that can take only a finite or countable number of distinct values, rather than a continuous range. Quantization arises when a system or process constrains possible values to discrete levels, whether by fundamental physical principles or by design choices in measurement and computation.

In physics, many quantities are quantized. Energy levels in atoms are discrete, and the energy of a photon is proportional to its frequency (E = h f). In quantum mechanics, observables often have discrete eigenvalues in a given system. Some phenomena, such as electric charge in certain contexts, show quantization in the sense that charge is observed in multiples of a fundamental unit.

In signal processing, quantization refers to mapping a continuous amplitude range to a finite set of levels during analog-to-digital conversion. Each sample is assigned to the nearest representable level, introducing quantization error or noise. The number of levels is determined by the bit depth; uniform quantization uses evenly spaced steps, while nonuniform quantization uses unequal steps to better match signal statistics.

In computing and applied mathematics, quantization also means discretization of a continuous domain for numerical methods and simulations. In modern hardware and AI, quantized representations—such as low-precision weights and activations—enable more efficient storage and faster computation, with some trade-offs in accuracy.

Overall, quantized describes anything that exists in or is represented by discrete units rather than a continuous spectrum.