paper-with-me

Papers

Exploring Multiple High-Scoring Subspaces in Generative Flow Networks

2026-02-12 · Xuan Yu, Xu Wang, Rui Zhu, Yudong Zhang, Yang Wang arxiv

As a probabilistic sampling framework, Generative Flow Networks (GFlowNets) show strong potential for constructing complex combinatorial objects through the sequential composition of elementary components. However, existing GFlowNets often suffer from excessive exploration over vast state spaces, leading to over-sampling of low-reward regions and convergence to suboptimal distributions. Effectively biasing GFlowNets toward high-reward solutions remains a non-trivial challenge. In this paper, we propose CMAB-GFN, which integrates a combinatorial multi-armed bandit (CMAB) framework with GFlowNet policies. The CMAB component prunes low-quality actions, yielding compact high-scoring subspaces for exploration. Restricting GFNs to these compact high-scoring subspaces accelerates the discovery of high-value candidates, while the exploration of different subspaces ensures that diversity is not sacrificed. Experimental results on multiple tasks demonstrate that CMAB-GFN generates higher-reward candidates than existing approaches.

📄 PDF Abstract BibTeX arXiv:2602.11491

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Sub-GAN: An Unsupervised Generative Model via Subspaces

2018-09-01 · ECCV 2018 9 · Jie Liang, Jufeng Yang, Hsin-Ying Lee, Kai Wang 외

The recent years have witnessed significant growth in constructing robust generative models to capture informative distributions of natural data. However, it is difficult to fully exploit the distribution of complex data…

Generative Adversarial Network

Exploring the Potential of Large Multimodal Models as Effective Alternatives for Pronunciation Assessment

2025-03-14 · Ke Wang, Lei He, Kun Liu, Yan Deng 외

Large Multimodal Models (LMMs) have demonstrated exceptional performance across a wide range of domains. This paper explores their potential in pronunciation assessment tasks, with a particular focus on evaluating the ca…

Pursuit of a Discriminative Representation for Multiple Subspaces via Sequential Games

2022-06-18 · Druv Pai, Michael Psenka, Chih-Yuan Chiu, Manxi Wu 외

We consider the problem of learning discriminative representations for data in a high-dimensional space with distribution supported on or around multiple low-dimensional linear subspaces. That is, we wish to compute a li…

Representation Learning

Learning Where To Look -- Generative NAS is Surprisingly Efficient

2022-03-16 · Jovita Lukasik, Steffen Jung, Margret Keuper

The efficient, automated search for well-performing neural architectures (NAS) has drawn increasing attention in the recent past. Thereby, the predominant research objective is to reduce the necessity of costly evaluatio…

Neural Architecture Search

Adversarial Subspace Generation for Outlier Detection in High-Dimensional Data

2025-04-10 · Jose Cribeiro-Ramallo, Federico Matteucci, Paul Enciu, Alexander Jenke 외

Outlier detection in high-dimensional tabular data is challenging since data is often distributed across multiple lower-dimensional subspaces -- a phenomenon known as the Multiple Views effect (MV). This effect led to a …

feature selectionOne-Class ClassificationOutlier DetectionStochastic Optimization