Research map / Probabilistic and scientific modeling / Dynamical systems
Bayesian Learning of Dynamic Multilayer Networks
Daniele Durante, Nabanita Mukherjee, Rebecca C. Steorts
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 |
|---|---|---|---|
| add layer-shared and layer-specific latent coordinates (Joint) instead of separate per-layer analyses (ablation) | dynamic multilayer latent space model · Simulation edge probability concentration | squared posterior bias: 0{.}016 -> 0{.}009 | Improved |
| add layer-shared and layer-specific latent coordinates (Joint) instead of separate per-layer analyses (ablation) | dynamic multilayer latent space model · Simulation edge probability concentration | posterior variance: 0{.}018 -> 0{.}011 | Improved |
| add layer-specific coordinates (Joint) instead of collapsed common coordinates (ablation) | dynamic multilayer latent space model · Simulation edge probability concentration | squared posterior bias: 0{.}016 -> 0{.}009 | Improved |
| add layer-specific coordinates (Joint) instead of collapsed common coordinates (ablation) | dynamic multilayer latent space model · Simulation edge probability concentration | posterior variance: 0{.}007 -> 0{.}011 | Hurt |
| add layer-specific coordinates (Joint) instead of collapsed common coordinates (ablation) | dynamic multilayer latent space model · Simulation edge probability concentration | posterior concentration: 0{.}023 -> 0{.}020 | Improved |
| increase number of actors V or number of times n | Joint model · Simulation edge probability concentration | posterior concentration: 0{.}020 -> 0{.}028 | Mixed |
| add information sharing across days (Joint) instead of separate analysis (ablation) | dynamic multilayer latent space model · Face-to-face contact data, out-of-sample, 10 coordinates | AUC: 0.83 -> 0.95 | Improved |
| add layer-specific coordinates (Joint) instead of collapsed (ablation) | dynamic multilayer latent space model · Face-to-face contact data, in-sample, 10 coordinates | AUC: 0.94 -> 0.97 | Improved |
| add layer-specific coordinates (Joint) instead of collapsed (ablation) | dynamic multilayer latent space model · Face-to-face contact data, out-of-sample, 10 coordinates | AUC: 0.95 -> 0.95 | No clear effect |
| vary number of latent coordinates from 8 to 12 | Joint model · Face-to-face contact data | AUC: 0.97 -> 0.97 | No clear effect |
| add information sharing across days (Joint) instead of separate analysis (ablation) | dynamic multilayer latent space model · Face-to-face contact data, out-of-sample, 10 coordinates | accuracy: 89% -> 94% | Improved |
| add layer-specific coordinates (Joint) instead of collapsed (ablation) | dynamic multilayer latent space model · Face-to-face contact data, in-sample, 10 coordinates | accuracy: 92% -> 95% | Improved |
Closest papers by meaning
- An Eigenmodel for Dynamic Multilayer Networks (JMLR 2023 · similarity 0.92)
- Nonparametric Network Models for Link Prediction (JMLR 2016 · similarity 0.88)
- A Flexible Latent Space Model for Multilayer Networks (ICML 2020 · similarity 0.88)
- A Poisson Gamma Probabilistic Model for Latent Node-Group Memberships in Dynamic Networks (AAAI 2018 · similarity 0.87)
- Dependent Relational Gamma Process Models for Longitudinal Networks (ICML 2018 · similarity 0.87)
- Modelling Populations of Interaction Networks via Distance Metrics (JMLR 2025 · similarity 0.87)
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.
