Ax

Ax is an accessible, general-purpose platform for understanding, managing, deploying, and automating adaptive experiments. Adaptive experimentation is the machine-learning guided process of iteratively exploring a (possibly infinite) parameter space in order to identify optimal configurations in a resource-efficient manner. Ax currently supports Bayesian optimization and bandit optimization as exploration strategies. Bayesian optimization in Ax is powered by BoTorch, a modern library for Bayesian optimization research built on PyTorch.

homepage: https://ax.dev/

version	toolchain
0.3.3	foss/2022a

(quick links: (all) - 0 - a - b - c - d - e - f - g - h - i - j - k - l - m - n - o - p - q - r - s - t - u - v - w - x - y - z)