Adaptive Rejection Sampler

Overview

This skill provides guidance for implementing Adaptive Rejection Sampling (ARS) algorithms. ARS is a method for generating samples from log-concave probability distributions by constructing piecewise linear upper and lower envelopes of the log-density function. The skill focuses on procedural approaches, performance optimization, and verification strategies rather than providing implementation code.

When to Use This Skill

Use this skill when:

• Implementing adaptive rejection sampling from scratch
• Working with log-concave distribution samplers
• Building statistical sampling algorithms that require envelope construction
• Debugging or optimizing existing ARS implementations
• The task involves R, Python, or other statistical computing environments

Implementation Approach

Phase 1: Algorithm Design Before Coding