On Model Selection Consistency of Lasso for High-Dimensional Ising Models
Xiangming Meng, Tomoyuki Obuchi, Yoshiyuki Kabashima

TL;DR
This paper provides a rigorous theoretical analysis of the model selection consistency of Lasso, with and without post-thresholding, for high-dimensional Ising models on various graph structures, matching prior conjectures and extending understanding.
Contribution
It rigorously proves the model selection consistency of Lasso for high-dimensional Ising models on regular and tree-like graphs, confirming previous non-rigorous conjectures and extending to post-thresholding methods.
Findings
Lasso without post-thresholding is consistent in the paramagnetic phase for RR graphs with sample complexity $n= ilde{O}(d^3\log p)$.
The same consistency result extends to general tree-like graphs under mild conditions.
Lasso with post-thresholding is consistent for general tree-like graphs without additional assumptions.
Abstract
We theoretically analyze the model selection consistency of least absolute shrinkage and selection operator (Lasso), both with and without post-thresholding, for high-dimensional Ising models. For random regular (RR) graphs of size with regular node degree and uniform couplings , it is rigorously proved that Lasso \textit{without post-thresholding} is model selection consistent in the whole paramagnetic phase with the same order of sample complexity as that of -regularized logistic regression (-LogR). This result is consistent with the conjecture in Meng, Obuchi, and Kabashima 2021 using the non-rigorous replica method from statistical physics and thus complements it with a rigorous proof. For general tree-like graphs, it is demonstrated that the same result as RR graphs can be obtained under mild assumptions of the dependency…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsMarkov Chains and Monte Carlo Methods · Statistical Methods and Inference · Stochastic processes and statistical mechanics
