Research map / Time series / Causal and generative time series
Forecasting Treatment Responses Over Time Using Recurrent Marginal Structural Networks
Bryan Lim
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 |
|---|---|---|---|
| replace MSM logistic IPTW with RNN propensity networks | R-MSN · Cancer simulation | one-step RMSE gamma=0: 80.9% / 78.7% | Improved |
| R-MSN versus MSM (linear IPTW model) | R-MSN · Cancer simulation | one-step RMSE gamma=0: 80.9% | Improved |
| R-MSN versus R-BTRC | R-MSN · Cancer simulation | one-step RMSE gamma=0: 66.1% | Improved |
| reduce confounding gamma from 10 to 0 | R-MSN · Cancer simulation | one-step normalized RMSE: 1.02% -> 0.92% | Improved |
| reduce confounding gamma from 10 to 0 | R-BTRC · Cancer simulation | one-step normalized RMSE: 2.09% -> 1.91% | Improved |
| seq2seq multi-step prediction vs MSM on training policy | R-MSN · Cancer simulation | multi-step normalized RMSE: +61% | Improved |
| seq2seq multi-step prediction vs MSM on unbiased policy | R-MSN · Cancer simulation | multi-step normalized RMSE: +95% | Improved |
| R-BTRC versus MSM on training policy | R-BTRC · Cancer simulation | multi-step normalized RMSE: -32% (RMSE up) | Hurt |
| R-BTRC versus MSM on unbiased assignment | R-BTRC · Cancer simulation | multi-step normalized RMSE: +66% | Improved |
| linear model with MSM IPTW versus plain linear | linear model · Cancer simulation | one-step RMSE | Improved |
Closest papers by meaning
- Estimating counterfactual treatment outcomes over time through adversarially balanced representations (ICLR 2020 · similarity 0.88)
- Time Series Deconfounder: Estimating Treatment Effects over Time in the Presence of Hidden Confounders (ICML 2020 · similarity 0.85)
- Data-Driven Discovery of Dynamical Systems in Pharmacology using Large Language Models (NeurIPS 2024 · similarity 0.84)
- LLM-Driven Treatment Effect Estimation Under Inference Time Text Confounding (NeurIPS 2025 · similarity 0.84)
- Risk Horizons: Structured Hypothesis Spaces for Longitudinal Clinical Prediction (NeurIPS 2026 · similarity 0.84)
- Stabilized Neural Prediction of Potential Outcomes in Continuous Time (ICLR 2025 · similarity 0.84)
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.
