# From Ambiguity to Action: A POMDP Perspective on Partial Multi-Label Ambiguity and Its Horizon-One Resolution

arXiv: https://arxiv.org/abs/2602.04255 · HTML: https://mlsanity.com/paper/2602.04255/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL training of language and agent models, one of 15 directions in Deep reinforcement learning methods.
- Builds on 8 of its 34 references that are in the map.
- First posted on arXiv 4 Feb 2026.

## Placed in

- Deep reinforcement learning methods / [RL training of language and agent models](https://mlsanity.com/deeprl/rl-training-of-language-and-agent-models/)

## Builds on (most cited first, 8 of 8)

- [Learning a Deep ConvNet for Multi-label Classification with Partial Labels](https://mlsanity.com/paper/1902.09720/) (2019)
- [Partial Label Learning with Self-Guided Retraining](https://mlsanity.com/paper/1902.03045/) (2019)
- [POMDP inference and robust solution via deep reinforcement learning: An application to railway optimal maintenance](https://mlsanity.com/paper/2307.08082/) (2023)
- [PLM: Partial Label Masking for Imbalanced Multi-label Classification](https://mlsanity.com/paper/2105.10782/) (2021)
- [Full or Weak annotations? An adaptive strategy for budget-constrained annotation campaigns](https://mlsanity.com/paper/2303.11678/) (2023)
- [Weakly Supervised Attended Object Detection Using Gaze Data as Annotations](https://mlsanity.com/paper/2204.07090/) (2022)
- [Reconsidering Feature Structure Information and Latent Space Alignment in Partial Multi-label Feature Selection](https://mlsanity.com/paper/2503.10115/) (2025)
- [Noise-Resistant Label Reconstruction Feature Selection for Partial Multi-Label Learning](https://mlsanity.com/paper/2506.04669/) (2025)

## Abstract (arXiv)

In partial multi-label learning (PML), the true labels are unobserved, which makes label disambiguation important but difficult. A key challenge is that ambiguous candidate labels can propagate errors into downstream tasks such as feature engineering. To solve this issue, we jointly model the disambiguation and feature selection tasks as Partially Observable Markov Decision Processes (POMDP) to turn PML risk minimization into expected-return maximization. Stage 1 trains a transformer policy via reinforcement learning to produce high-quality hard pseudo-labels; Stage 2 describes feature selection as a sequential reinforcement learning problem, selecting features step by step and outputting an interpretable global ranking. We further provide the theoretical analysis of PML-POMDP correspondence and the excess-risk bound that decompose the error into pseudo label quality term and sample size. Experiments in multiple metrics and data sets verify the advantages of the framework.