Research map / Learning theory and algorithms / Matrix and tensor methods
Logistic Regression Under Network Dependence
Somabha Mukherjee, Ziang Niu, Sagnik Halder, Bhaswar B. Bhattacharya, George Michailidis
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Model network dependence with Ising interaction term instead of plain penalized logistic regression | PMPL · Synthetic ER/SBM/beta-model/preferential attachment, d=50, N up to 1200 | l1 estimation error | Improved |
| Model network dependence with Ising interaction term instead of plain penalized logistic regression | PMPL · Synthetic networks, d=50 | l2 estimation error | Mixed |
| Increase network size N from 200 to 1200 | PMPL · Synthetic networks, d=50 | estimation error | Improved |
| Increase network density c | PMPL · Erdos-Renyi G(N,c/N), N=600 | estimation error | Hurt |
| Add network interaction term (Ising) versus logistic regression: variance of errors | PMPL · Synthetic networks | variance of estimation error | Improved |
| Increase nearest-neighbor graph K from 1 to 2 (ablation) | PMPL · Visium breast cancer gene selection | selected genes: same 6 genes | No clear effect |
| Increase nearest-neighbor graph K from 2 to 3 (ablation) | PMPL · Visium breast cancer gene selection | selected genes: SPP1 no longer selected | Mixed |
| Increase K from 1 to 3 (estimated dependence beta) (ablation) | PMPL · Visium breast cancer | estimated beta: 0.1203 -> 0.2870 | Improved |
| Use only 6 PMPL-selected genes instead of top 50 variable genes for clustering | Leiden clustering · Visium breast cancer | clustering agreement | No clear effect |
Closest papers by meaning
- Embedding Network Autoregression for Time Series Analysis and Causal Peer Effect Inference (JMLR 2026 · similarity 0.89)
- Using Embeddings for Causal Estimation of Peer Influence in Social Networks (NeurIPS 2022 · similarity 0.88)
- Inference for Network Regression Models with Community Structure (ICML 2021 · similarity 0.87)
- Telling Peer Direct Effects from Indirect Effects in Observational Network Data (ICML 2025 · similarity 0.87)
- A Generic Family of Graphical Models: Diversity, Efficiency, and Heterogeneity (ICML 2025 · similarity 0.85)
- Gene Regulatory Network Inference in the Presence of Selection Bias and Latent Confounders (NeurIPS 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
