mlsanityData to 14 Oct 2026

Stochastic Encodings for Active Feature Acquisition

arXiv 2508.01957 ↗First posted 3 Aug 2025cs.LG stat.MLcited by 9

Not on the map. Why?

Abstract (arXiv)

Active Feature Acquisition is an instance-wise, sequential decision making problem. The aim is to dynamically select which feature to measure based on current observations, independently for each test instance. Common approaches either use Reinforcement Learning, which experiences training difficulties, or greedily maximize the conditional mutual information of the label and unobserved features, which makes myopic acquisitions. To address these shortcomings, we introduce a latent variable model, trained in a supervised manner. Acquisitions are made by reasoning about the features across many possible unobserved realizations in a stochastic latent space. Extensive evaluation on a large range of synthetic and real datasets demonstrates that our approach reliably outperforms a diverse set of baselines.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Builds on 29 of its 77 references that are in the map. First posted on arXiv 3 Aug 2025.

Builds on

29 of 77 references.

References · 29

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 7,005
  3. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,349
  4. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
  5. 2016Deep Variational Information Bottleneckcited by 930
  6. 2017Deep Reinforcement Learning that Matterscited by 754
  7. 2014Recurrent Models of Visual Attentioncited by 349
  8. 2018Deep Reinforcement Learning and the Deadly Triadcited by 174
  9. 2018EDDI: Efficient Dynamic Discovery of High-Value Information with Partial VAEcited by 87
  10. 2023Learning to Maximize Mutual Information for Dynamic Feature Selectioncited by 56
  11. 2020Active Feature Acquisition with Generative Surrogate Modelscited by 54
  12. 2017Classification with Costly Features using Deep Reinforcement Learningcited by 34
17 more
  1. 2019Flow Models for Arbitrary Conditional Likelihoodscited by 30
  2. 2022Interpretable by Design: Learning Predictors by Composing Interpretable Queriescited by 28
  3. 2023Variational Information Pursuit for Interpretable Predictionscited by 25
  4. 2020Reinforcement Learning with Efficient Active Feature Acquisitioncited by 21
  5. 2023Estimating Conditional Mutual Information for Dynamic Feature Selectioncited by 21
  6. 2019Opportunistic Learning: Budgeted Cost-Sensitive Learning from Data Streamscited by 20
  7. 2022Missing Data Imputation and Acquisition with Deep Hierarchical Models and Hamiltonian Monte Carlocited by 19
  8. 2023Acquisition Conditioned Oracle for Nongreedy Active Feature Acquisitioncited by 19
  9. 2016Pruning Random Forests for Prediction on a Budgetcited by 14
  10. 2020BSODA: A Bipartite Scalable Framework for Online Disease Diagnosiscited by 14
  11. 2022Active Acquisition for Multimodal Temporal Data: A Challenging Decision-Making Taskcited by 13
  12. 2021A Probabilistic Hard Attention Model For Sequentially Observed Scenescited by 12
  13. 2016Active Information Acquisitioncited by 11
  14. 2018Learning Sparse Latent Representations with the Deep Copula Information Bottleneckcited by 9
  15. 2015Feature-Budgeted Random Forestcited by 6
  16. 2018Dynamic Feature Acquisition Using Denoising Autoencoderscited by 6
  17. 2020Dynamic Feature Acquisition with Arbitrary Conditional Flowscited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 7.

9 papers.

  1. 2025Learning-To-Measure: In-Context Active Feature Acquisitioncited by 5
  2. 2025Information Templates: A New Paradigm for Intelligent Active Feature Acquisitioncited by 3
  3. 2026Non-Myopic Active Feature Acquisition via Pathwise Policy Gradientscited by 1
  4. 2026Relaxed Efficient Acquisition of Context and Temporal Featurescited by 0
  5. 2026Hypernetworks for Dynamic Feature Selectioncited by 0
  6. 2026BRiG-AFA: Bellman Risk-to-Go Learning for Non-Myopic Active Feature Acquisitioncited by 0
  7. 2026Learning Source Acquisition Policies by Offline Planningcited by 0
  8. 2026Active Feature Acquisition With Incomplete Training Datacited by 0
  9. 2026Measure Less, Know More: Self-Supervised Test-Time Feature Acquisitioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025AFABench: A Generic Framework for Benchmarking Active Feature Acquisition
  2. 2023Evaluation of Active Feature Acquisition Methods for Time-varying Feature Settings
  3. 2026Sampling-guided exploration of active feature selection policies
  4. 2023Evaluation of Active Feature Acquisition Methods for Static Feature Settings
  5. 2026AFA-BANDIT: Provably Near-Optimal Online Multi-Feature Classification Under Budget Constraints
  6. 2024Knockoff-Guided Feature Selection via A Single Pre-trained Reinforced Agent
  7. 2026Adaptive Active Learning for Regression via Reinforcement Learning
  8. 2025Optimizing Active Learning in Vision-Language Models via Parameter-Efficient Uncertainty Calibration