Research map / Learning theory and algorithms / Prediction and decision losses
Learning Influence Functions from Incomplete Observations
Xinran He, Ke Xu, David Kempe, Yan Liu
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 |
|---|---|---|---|
| Account for missing activations rather than treating incomplete cascades as complete | Influence function learning · MemeTracker (seven groups) | MAE: 18% reduction in MAE | Improved |
| Account for missing activations | Influence function learning · Synthetic cascades at retention rates as small as 0.5 | MAE | Improved |
| Use an incorrect retention rate (ablation) | Incomplete-observation-aware learner · Synthetic cascades generated with true retention rate 0.8 | MAE | Hurt |
| Use a retention rate of 1 instead of the true rate 0.8 (ablation) | Incomplete-observation-aware learner · Synthetic cascades | MAE | Hurt |
| Allow independently perturbed node-specific retention rates around the estimated mean (ablation) | Incomplete-observation-aware learner · Synthetic cascades, mean retention rate 0.8 | Relative MAE difference | No clear effect |
| Fit influence using logistic regression | Influence function learning · Synthetic cascades | MAE | Hurt |
| Fit influence using linear regression | Influence function learning · Synthetic cascades | MAE | Hurt |
| Use CIC inference instead of DIC inference under model misspecification | Parameter-fitting approaches · Synthetic cascades | MAE | Improved |
Closest papers by meaning
- Efficient Online Influence Maximization under the Independent Cascade Model with Node-Level Feedback (ICML 2026 · similarity 0.87)
- Learning Influence Adoption in Heterogeneous Networks (AAAI 2022 · similarity 0.86)
- Online Influence Maximization under Linear Threshold Model (NeurIPS 2020 · similarity 0.86)
- The Importance of Communities for Learning to Influence (NeurIPS 2017 · similarity 0.86)
- DiffIM: Differentiable Influence Minimization with Surrogate Modeling and Continuous Relaxation (AAAI 2025 · similarity 0.85)
- Efficient PAC Learnability of Dynamical Systems Over Multilayer Networks (ICML 2024 · similarity 0.83)
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.
