# Online Feedback Efficient Active Target Discovery in Partially Observable Environments

arXiv: https://arxiv.org/abs/2505.06535 · HTML: https://mlsanity.com/paper/2505.06535/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 36 references that are in the map.
- First posted on arXiv 10 May 2025.

## Builds on (most cited first, 20 of 20)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Diffusion Posterior Sampling for General Noisy Inverse Problems](https://mlsanity.com/paper/2209.14687/) (2022)
- [MaskGAN: Towards Diverse and Interactive Facial Image Manipulation](https://mlsanity.com/paper/1907.11922/) (2019)
- [DOTA: A Large-scale Dataset for Object Detection in Aerial Images](https://mlsanity.com/paper/1711.10398/) (2017)
- [Solving Linear Inverse Problems Provably via Posterior Sampling with Latent Diffusion Models](https://mlsanity.com/paper/2307.00619/) (2023)
- [A Patient-Centric Dataset of Images and Metadata for Identifying Melanomas Using Clinical Context](https://mlsanity.com/paper/2008.07360/) (2020)
- [Learning When and Where to Zoom with Deep Reinforcement Learning](https://mlsanity.com/paper/2003.00425/) (2020)
- [Experimental design for MRI by greedy policy search](https://mlsanity.com/paper/2010.16262/) (2020)
- [Learning to Look Around: Intelligently Exploring Unseen Environments for Unknown Tasks](https://mlsanity.com/paper/1709.00507/) (2017)
- [Look-ahead before you leap: end-to-end active recognition by forecasting the effect of motion](https://mlsanity.com/paper/1605.00164/) (2016)
- [Consistency driven Sequential Transformers Attention Model for Partially Observable Scenes](https://mlsanity.com/paper/2204.00656/) (2022)
- [GOMAA-Geo: GOal Modality Agnostic Active Geo-localization](https://mlsanity.com/paper/2406.01917/) (2024)
- [A Visual Active Search Framework for Geospatial Exploration](https://mlsanity.com/paper/2211.15788/) (2022)
- [A Partially Supervised Reinforcement Learning Framework for Visual Active Search](https://mlsanity.com/paper/2310.09689/) (2023)
- [Active Diffusion Subsampling](https://mlsanity.com/paper/2406.14388/) (2024)
- [Accelerated Intravascular Ultrasound Imaging using Deep Reinforcement Learning](https://mlsanity.com/paper/2201.09522/) (2022)
- [Amortized nonmyopic active search via deep imitation learning](https://mlsanity.com/paper/2405.15031/) (2024)

## Built on by (most cited first, 1 of 1)

- [Active Target Discovery under Uninformative Prior: The Power of Permanent and Transient Memory](https://mlsanity.com/paper/2510.16676/) (2025)

## Abstract (arXiv)

In various scientific and engineering domains, where data acquisition is costly--such as in medical imaging, environmental monitoring, or remote sensing--strategic sampling from unobserved regions, guided by prior observations, is essential to maximize target discovery within a limited sampling budget. In this work, we introduce Diffusion-guided Active Target Discovery (DiffATD), a novel method that leverages diffusion dynamics for active target discovery. DiffATD maintains a belief distribution over each unobserved state in the environment, using this distribution to dynamically balance exploration-exploitation. Exploration reduces uncertainty by sampling regions with the highest expected entropy, while exploitation targets areas with the highest likelihood of discovering the target, indicated by the belief distribution and an incrementally trained reward model designed to learn the characteristics of the target. DiffATD enables efficient target discovery in a partially observable environment within a fixed sampling budget, all without relying on any prior supervised training. Furthermore, DiffATD offers interpretability, unlike existing black--box policies that require extensive supervised training. Through extensive experiments and ablation studies across diverse domains, including medical imaging, species discovery, and remote sensing, we show that DiffATD performs significantly better than baselines and competitively with supervised methods that operate under full environmental observability.