activity
20242026
collaborators

6 papers

stat.CO2026

Hessian-informed, Coordinate Friendly Hamiltonian Monte Carlo in Linear Time

Son Luu, Nikola Surjanovic, Zuheng Xu +2

Riemannian Hamiltonian Monte Carlo (RHMC) is a promising MCMC methodology thanks to its ability to accommodate position-dependent preconditioning and multi-step proposals. While RH…

cs.LG2025

AutoGD: Automatic Learning Rate Selection for Gradient Descent

Nikola Surjanovic, Alexandre Bouchard-Côté, Trevor Campbell

The performance of gradient-based optimization methods, such as standard gradient descent (GD), greatly depends on the choice of learning rate. However, it can require a non-trivia…

cs.LG2025

AutoSGD: Automatic Learning Rate Selection for Stochastic Gradient Descent

Nikola Surjanovic, Alexandre Bouchard-Côté, Trevor Campbell

The learning rate is an important tuning parameter for stochastic gradient descent (SGD) and can greatly influence its performance. However, appropriate selection of a learning rat…

stat.CO2025

AutoStep: Locally adaptive involutive MCMC

Tiange Liu, Nikola Surjanovic, Miguel Biron-Lattes +2

Many common Markov chain Monte Carlo (MCMC) kernels can be formulated using a deterministic involutive proposal with a step size parameter. Selecting an appropriate step size is of…

stat.CO2025

Is Gibbs sampling faster than Hamiltonian Monte Carlo on GLMs?

Son Luu, Zuheng Xu, Nikola Surjanovic +3

The Hamiltonian Monte Carlo (HMC) algorithm is often lauded for its ability to effectively sample from high-dimensional distributions. In this paper we challenge the presumed domin…

stat.CO2024

Uniform Ergodicity of Parallel Tempering With Efficient Local Exploration

Nikola Surjanovic, Saifuddin Syed, Alexandre Bouchard-Côté +1

Non-reversible parallel tempering (NRPT) is an effective algorithm for sampling from target distributions with complex geometry, such as those arising from posterior distributions…