kld

KLD is an acronym with multiple meanings across different fields. The most common references are in information theory, where it denotes the Kullback–Leibler divergence, and in computing, where it stands for kernel loadable modules used by some Unix-like operating systems.

The Kullback–Leibler divergence, D_KL(P||Q), measures the expected log difference between two probability distributions P and Q. For discrete distributions, it is defined as D_KL(P||Q) = sum_x P(x) log(P(x)/Q(x)). For continuous distributions, the sum is replaced by an integral. D_KL is non-negative and equals zero if and only if P and Q coincide almost everywhere. It is not symmetric, so D_KL(P||Q) generally differs from D_KL(Q||P). The divergence is widely used in model evaluation, information theory, statistics, and machine learning, including in variational inference and optimization problems.

In FreeBSD and other Unix-like systems, a KLD refers to a kernel loadable module: a piece of code that can be dynamically loaded into the operating system kernel to extend functionality, such as drivers or filesystem support. The kernel module framework provides utilities to manage these modules at runtime, including commands like kldload, kldunload, and kldstat. This dynamic loading capability allows optional features to be added or removed without rebooting the system.