Discussion Lead: Bob Carpenter [CCM]
Topic: GIST: Gibbs self-tuning for locally adaptive HMC
Link: TBD
Abstract: We present a novel and flexible framework for localized tuning of Hamiltonian Monte Carlo samplers by sampling an algorithm's tuning parameters conditionally based on the position and momentum at each step. For adaptively sampling path lengths, we show that randomized Hamiltonian Monte Carlo, the no U-turn sampler, and the apogee-to-apogee path Sampler all fit within this unified framework as special cases. The framework is illustrated with a simple alternative to NUTS for locally adapting path lengths.