# Context-Action Embedding Learning for Off-Policy Evaluation in Contextual Bandits

arXiv: https://arxiv.org/abs/2509.00648 · HTML: https://mlsanity.com/paper/2509.00648/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 23 references that are in the map.
- First posted on arXiv 31 Aug 2025.

## Builds on (most cited first, 12 of 12)

- [Doubly Robust Policy Evaluation and Optimization](https://mlsanity.com/paper/1503.02834/) (2015)
- [More Robust Doubly Robust Off-policy Evaluation](https://mlsanity.com/paper/1802.03493/) (2018)
- [Optimal and Adaptive Off-policy Evaluation in Contextual Bandits](https://mlsanity.com/paper/1612.01205/) (2016)
- [Doubly robust off-policy evaluation with shrinkage](https://mlsanity.com/paper/1907.09623/) (2019)
- [Policy Evaluation and Optimization with Continuous Treatments](https://mlsanity.com/paper/1802.06037/) (2018)
- [Off-Policy Evaluation for Large Action Spaces via Embeddings](https://mlsanity.com/paper/2202.06317/) (2022)
- [Learning from eXtreme Bandit Feedback](https://mlsanity.com/paper/2009.12947/) (2020)
- [Off-Policy Evaluation of Slate Bandit Policies via Optimizing Abstraction](https://mlsanity.com/paper/2402.02171/) (2024)
- [Off-Policy Evaluation for Large Action Spaces via Policy Convolution](https://mlsanity.com/paper/2310.15433/) (2023)
- [Learning Action Embeddings for Off-Policy Evaluation](https://mlsanity.com/paper/2305.03954/) (2023)
- [Local Metric Learning for Off-Policy Evaluation in Contextual Bandits with Continuous Actions](https://mlsanity.com/paper/2210.13373/) (2022)
- [Clustering Context in Off-Policy Evaluation](https://mlsanity.com/paper/2502.21304/) (2025)

## Abstract (arXiv)

We consider off-policy evaluation (OPE) in contextual bandits with finite action space. Inverse Propensity Score (IPS) weighting is a widely used method for OPE due to its unbiased, but it suffers from significant variance when the action space is large or when some parts of the context-action space are underexplored. Recently introduced Marginalized IPS (MIPS) estimators mitigate this issue by leveraging action embeddings. However, these embeddings do not minimize the mean squared error (MSE) of the estimators and do not consider context information. To address these limitations, we introduce Context-Action Embedding Learning for MIPS, or CAEL-MIPS, which learns context-action embeddings from offline data to minimize the MSE of the MIPS estimator. Building on the theoretical analysis of bias and variance of MIPS, we present an MSE-minimizing objective for CAEL-MIPS. In the empirical studies on a synthetic dataset and a real-world dataset, we demonstrate that our estimator outperforms baselines in terms of MSE.