# Evaluation of Active Feature Acquisition Methods for Time-varying Feature Settings

arXiv: https://arxiv.org/abs/2312.01530 · HTML: https://mlsanity.com/paper/2312.01530/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Offline RL algorithms, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 22 of its 68 references that are in the map.
- First posted on arXiv 3 Dec 2023.

## Placed in

- Offline RL and imitation learning from data / [Offline RL algorithms](https://mlsanity.com/offlinerl/offline-rl-algorithms/)

## Builds on (most cited first, 22 of 22)

- [Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems](https://mlsanity.com/paper/2005.01643/) (2020)
- [Data-Efficient Off-Policy Policy Evaluation for Reinforcement Learning](https://mlsanity.com/paper/1604.00923/) (2016)
- [Semiparametric doubly robust targeted double machine learning: a review](https://mlsanity.com/paper/2203.06469/) (2022)
- [Double Reinforcement Learning for Efficient Off-Policy Evaluation in Markov Decision Processes](https://mlsanity.com/paper/1908.08526/) (2019)
- [EDDI: Efficient Dynamic Discovery of High-Value Information with Partial VAE](https://mlsanity.com/paper/1809.11142/) (2018)
- [What's a good imputation to predict with missing values?](https://mlsanity.com/paper/2106.00311/) (2021)
- [Active Feature Acquisition with Generative Surrogate Models](https://mlsanity.com/paper/2010.02433/) (2020)
- [Full Law Identification In Graphical Models Of Missing Data: Completeness Results](https://mlsanity.com/paper/2004.04872/) (2020)
- [Classification with Costly Features as a Sequential Decision-Making Problem](https://mlsanity.com/paper/1909.02564/) (2019)
- [Reinforcement Learning with Efficient Active Feature Acquisition](https://mlsanity.com/paper/2011.00825/) (2020)
- [Identification In Missing Data Models Represented By Directed Acyclic Graphs](https://mlsanity.com/paper/1907.00241/) (2019)
- [ASAC: Active Sensing using Actor-Critic models](https://mlsanity.com/paper/1906.06796/) (2019)
- [Active Feature Acquisition with Supervised Matrix Completion](https://mlsanity.com/paper/1802.05380/) (2018)
- [Inverse Active Sensing: Modeling and Understanding Timely Decision-Making](https://mlsanity.com/paper/2006.14141/) (2020)
- [Semiparametric theory](https://mlsanity.com/paper/1709.06418/) (2017)
- [An Optimal Policy for Patient Laboratory Tests in Intensive Care Units](https://mlsanity.com/paper/1808.04679/) (2018)
- [Dynamic Measurement Scheduling for Event Forecasting using Deep RL](https://mlsanity.com/paper/1901.09699/) (2019)
- [Dynamic Feature Acquisition with Arbitrary Conditional Flows](https://mlsanity.com/paper/2006.07701/) (2020)
- [Efficient estimation of optimal regimes under a no direct effect assumption](https://mlsanity.com/paper/1908.10448/) (2019)
- [Towards Robust Active Feature Acquisition](https://mlsanity.com/paper/2107.04163/) (2021)
- [From Single-Hospital to Multi-Centre Applications: Enhancing the Generalisability of Deep Learning Models for Adverse Event Prediction in the ICU](https://mlsanity.com/paper/2303.15354/) (2023)
- [Active Selection of Classification Features](https://mlsanity.com/paper/2102.13636/) (2021)

## Built on by (most cited first, 9 of 9)

- [A Survey on Active Feature Acquisition Strategies](https://mlsanity.com/paper/2502.11067/) (2025)
- [Cohort-Based Active Modality Acquisition](https://mlsanity.com/paper/2505.16791/) (2025)
- [Learning-To-Measure: In-Context Active Feature Acquisition](https://mlsanity.com/paper/2510.12624/) (2025)
- [Evaluation of Active Feature Acquisition Methods for Static Feature Settings](https://mlsanity.com/paper/2312.03619/) (2023)
- [Timely Clinical Diagnosis through Active Test Selection](https://mlsanity.com/paper/2510.18988/) (2025)
- [Characterizing the Predictive Impact of Modalities with Supervised Latent-Variable Modeling](https://mlsanity.com/paper/2602.16979/) (2026)
- [BRiG-AFA: Bellman Risk-to-Go Learning for Non-Myopic Active Feature Acquisition](https://mlsanity.com/paper/2608.02305/) (2026)
- [Counterfactual Evaluation of Temporal Observation Protocols](https://mlsanity.com/paper/2608.22221/) (2026)
- [When Does Test-Time Physical Diagnosis Pay? A Frozen Policy Buys Evidence It Never Reads](https://mlsanity.com/paper/2609.22299/) (2026)

## Abstract (arXiv)

Machine learning methods often assume that input features are available at no cost. However, in domains like healthcare, where acquiring features could be expensive or harmful, it is necessary to balance a feature's acquisition cost against its predictive value. The task of training an AI agent to decide which features to acquire is called active feature acquisition (AFA). By deploying an AFA agent, we effectively alter the acquisition strategy and trigger a distribution shift. To safely deploy AFA agents under this distribution shift, we present the problem of active feature acquisition performance evaluation (AFAPE). We examine AFAPE under i) a no direct effect (NDE) assumption, stating that acquisitions do not affect the underlying feature values; and ii) a no unobserved confounding (NUC) assumption, stating that retrospective feature acquisition decisions were only based on observed features. We show that one can apply missing data methods under the NDE assumption and offline reinforcement learning under the NUC assumption. When NUC and NDE hold, we propose a novel semi-offline reinforcement learning framework. This framework requires a weaker positivity assumption and introduces three new estimators: A direct method (DM), an inverse probability weighting (IPW), and a double reinforcement learning (DRL) estimator.