Research map / Probabilistic and scientific modeling / Treatment effects
Deep Proxy Causal Learning and its Application to Confounded Bandit Policy Evaluation
Liyuan Xu, Heishiro Kanagawa, Arthur Gretton
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 |
|---|---|---|---|
| use DFPV (neural adaptive features) instead of fixed-feature KPV/PMMR | proxy causal learning two-stage regression · Demand design structural function | out-of-sample MSE | Improved |
| use DFPV instead of KPV/PMMR with high-dimensional image treatment | proxy causal learning · dSprite structural function | out-of-sample MSE | Improved |
| use KPV instead of PMMR as data size grows | kernel PCL baselines · Demand design and dSprite | out-of-sample MSE | Mixed |
| use CEVAE (VAE confounder recovery with proxies) instead of kernel PCL methods | baseline · Demand design | out-of-sample MSE | Hurt |
| use CEVAE instead of kernel PCL methods | baseline · dSprite | out-of-sample MSE | Hurt |
| use DFPV instead of KPV/PMMR for policy depending on fuel costs | bandit off-policy evaluation · Confounded bandit OPE, cost policy | out-of-sample MSE | Improved |
| use DFPV instead of KPV/PMMR for policy depending on current price | bandit off-policy evaluation · Confounded bandit OPE, price policy | out-of-sample MSE | Improved |
Closest papers by meaning
- Density Ratio-Free Doubly Robust Proxy Causal Learning (NeurIPS 2025 · similarity 0.89)
- Coupling Generative Modeling and an Autoencoder with the Causal Bridge (NeurIPS 2025 · similarity 0.84)
- Transferring Causal Effects using Proxies (NeurIPS 2025 · similarity 0.84)
- Causal Inference with Conditional Front-Door Adjustment and Identifiable Variational Autoencoder (ICLR 2024 · similarity 0.84)
- Turning Sand to Gold: Recycling Data to Bridge On-Policy and Off-Policy Learning via Causal Bound (NeurIPS 2025 · similarity 0.84)
- BECAUSE: Bilinear Causal Representation for Generalizable Offline Model-based Reinforcement Learning (NeurIPS 2024 · 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.
