Mixing-Free and Signal-Optimal Learning of Gaussian Graphical Models from Glauber Dynamics
Vignesh Tirukkonda, Gautam Dasarathy
stat.ML, cs.LG, math.ST
Submitted: 2026-07-20
Comments: Minor correction to the proof of Lemma 15 (the statement and implications remain the same). For clarity and completeness: (1) separated pointwise recovery horizon guarantees from uniform ones, and (2) rewrote slow-mixing proposition (Proposition 2) over the class of problems with bounded initialization energy
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Entropy contraction of the Gibbs sampler under log-concavity
- Exponential Reduction in Sample Complexity with Learning of Ising Model Dynamics
- A Unified Approach to Learning Ising Models: Beyond Independence and Bounded Width
- Better Models and Algorithms for Learning Ising Models from Dynamics
- Learning Gaussian Graphical Models from a Glauber Trajectory Without Mixing
- Local and Mixing-Based Algorithms for Gaussian Graphical Model Selection from Glauber Dynamics
- A mixing time bound for Gibbs sampling from log-smooth log-concave distributions
Related papers
- Behavior of prediction performance metrics with rare events
- Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
- A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
- Scalable Posterior Uncertainty for Flexible Density-Based Clustering
- A Transport Geometry of Benign Overfitting
- Efficient quadratic entropy with distance sketches