English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

We consider the problem of learning the underlying graph of a sparse Isingmodel with $p$ nodes from $n$ i.i.d. samples. The most recent and bestperforming approaches combine an empirical loss (the logistic regression lossor the interaction screening loss) with a regularizer (an L1 penalty or an L1constraint). This results in a convex problem that can be solved separately foreach node of the graph. In this work, we leverage the cardinality constraint L0norm, which is known to properly induce sparsity, and further combine it withan L2 norm to better model the non-zero coefficients. We show that our proposedestimators achieve an improved sample complexity, both (a) theoretically, byreaching new state-of-the-art upper bounds for recovery guarantees, and (b)empirically, by showing sharper phase transitions between poor and fullrecovery for graph topologies studied in the literature, when compared to theirL1-based state-of-the-art methods.

Sample-Efficient L0-L2 Constrained Structure Learning of Sparse Ising Models