Research map / Time series / Causal and generative time series
Recovering the Graph Underlying Networked Dynamical Systems under Partial Observability: A Deep Learning Approach
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use CNNs on normalized lag-covariance feature vectors for pairwise edge classification | Partial-observability graph identification · Real-world and synthetic networks | accuracy | Helped |
| Use CNNs on normalized lag-covariance feature vectors for pairwise edge classification | Partial-observability graph identification · Structurally distinct networks | sample complexity | Helped |
| Train on a synthetic Erdős–Rényi network and apply the CNN to real-world networks | CNN graph classifier · Brain structural connectivity and enzyme biochemical network | accuracy | Helped |
| Vary the input noise variance while training at σ²=0.5 (ablation) | CNN and linear SVM · Noise-level robustness | accuracy | No clear effect |
| Add the partial-observability Granger estimator to the lag-covariance feature vector (ablation) | CNN graph classifier · Graph structure identification with |S|=20 observed nodes | accuracy | Helped |
Closest papers by meaning
- Learning the Causal Structure of Networked Dynamical Systems under Latent Nodes and Structured Noise (AAAI 2024 · similarity 0.93)
- A Graph Dynamics Prior for Relational Inference (AAAI 2024 · similarity 0.87)
- Graph Structure Inference with BAM: Neural Dependency Processing via Bilinear Attention (NeurIPS 2024 · similarity 0.86)
- Out-of-Domain Generalization in Dynamical Systems Reconstruction (ICML 2024 · similarity 0.86)
- When Graph Neural Networks Meet Dynamic Mode Decomposition (ICLR 2025 · similarity 0.86)
- A Dynamical Systems-Inspired Pruning Strategy for Addressing Oversmoothing in Graph Attention Networks (ICML 2025 · similarity 0.86)
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.