AdaMuS: Adaptive Multi-view Sparsity Learning for Dimensionally Unbalanced Data

• URL: http://arxiv.org/abs/2603.17610v1
• Date: Wed, 18 Mar 2026 11:22:47 GMT
• Title: AdaMuS: Adaptive Multi-view Sparsity Learning for Dimensionally Unbalanced Data
• Abstract summary: In emotion recognition tasks, video frames often reach dimensions of $106$, while physiological signals comprise only $101$ dimensions.<n>Existing methods often bias towards high-dimensional data, overlooking the low-dimensional views.<n>We propose a sparse fusion paradigm that flexibly suppresses redundant dimensions and effectively aligns each view.
• Score: 25.19474314673427
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Multi-view learning primarily aims to fuse multiple features to describe data comprehensively. Most prior studies implicitly assume that different views share similar dimensions. In practice, however, severe dimensional disparities often exist among different views, leading to the unbalanced multi-view learning issue. For example, in emotion recognition tasks, video frames often reach dimensions of $10^6$, while physiological signals comprise only $10^1$ dimensions. Existing methods typically face two main challenges for this problem: (1) They often bias towards high-dimensional data, overlooking the low-dimensional views. (2) They struggle to effectively align representations under extreme dimensional imbalance, which introduces severe redundancy into the low-dimensional ones. To address these issues, we propose the Adaptive Multi-view Sparsity Learning (AdaMuS) framework. First, to prevent ignoring the information of low-dimensional views, we construct view-specific encoders to map them into a unified dimensional space. Given that mapping low-dimensional data to a high-dimensional space often causes severe overfitting, we design a parameter-free pruning method to adaptively remove redundant parameters in the encoders. Furthermore, we propose a sparse fusion paradigm that flexibly suppresses redundant dimensions and effectively aligns each view. Additionally, to learn representations with stronger generalization, we propose a self-supervised learning paradigm that obtains supervision information by constructing similarity graphs. Extensive evaluations on a synthetic toy dataset and seven real-world benchmarks demonstrate that AdaMuS consistently achieves superior performance and exhibits strong generalization across both classification and semantic segmentation tasks.

Related papers

• Learning Subspace-Preserving Sparse Attention Graphs from Heterogeneous Multiview Data [23.128540151031846]
  We propose a sparse attention graph learning (SAGL) method that learns subspace-preserving sparse attention graphs from heterogeneous multiview data.<n>SAGL consistently outperforms the state-of-the-art unsupervised transfer learning approaches.
  arXiv  Detail & Related papers  (2026-05-12T09:56:28Z)
• VIEW2SPACE: Studying Multi-View Visual Reasoning from Sparse Observations [47.94531550391802]
  Multi-view visual reasoning is essential for intelligent systems that must understand complex environments from sparse and discrete viewpoints.<n>In real-world scenarios, reasoning across views requires integrating partial observations without explicit guidance.<n>We leverage physically grounded simulation to construct diverse, high-fidelity 3D scenes with precise per-view metadata.
  arXiv  Detail & Related papers  (2026-03-17T13:36:30Z)
• Revisiting Multi-Task Visual Representation Learning [52.93947931352643]
  We introduce MTV, a principled multi-task visual pretraining framework.<n>We leverage high-capacity "expert" models to synthesize dense, structured pseudo-labels at scale.<n>Our results demonstrate that MTV achieves "best-of-both-worlds" performance.
  arXiv  Detail & Related papers  (2026-01-20T11:59:19Z)
• Consensus dimension reduction via multi-view learning [0.6015898117103067]
  We advocate for a consensus approach, which summarizes multiple visualizations into a single consensus dimension reduction visualization.<n>Here, we leverage ideas from multi-view learning in order to identify the patterns that are most stable or shared across the many different dimension reduction visualizations.<n>We demonstrate that this consensus visualization effectively identifies and preserves the shared low-dimensional data structure through both simulated and real-world case studies.
  arXiv  Detail & Related papers  (2025-12-16T22:32:58Z)
• 360 Layout Estimation via Orthogonal Planes Disentanglement and Multi-view Geometric Consistency Perception [56.84921040837699]
  Existing panoramic layout estimation solutions tend to recover room boundaries from a vertically compressed sequence, yielding imprecise results. We propose an orthogonal plane disentanglement network (termed DOPNet) to distinguish ambiguous semantics. We also present an unsupervised adaptation technique tailored for horizon-depth and ratio representations. Our solution outperforms other SoTA models on both monocular layout estimation and multi-view layout estimation tasks.
  arXiv  Detail & Related papers  (2023-12-26T12:16:03Z)
• Hierarchical Mutual Information Analysis: Towards Multi-view Clustering in The Wild [9.380271109354474]
  This work proposes a deep MVC framework where data recovery and alignment are fused in a hierarchically consistent way to maximize the mutual information among different views. To the best of our knowledge, this could be the first successful attempt to handle the missing and unaligned data problem separately with different learning paradigms.
  arXiv  Detail & Related papers  (2023-10-28T06:43:57Z)
• Mitigating Representation Bias in Action Recognition: Algorithms and Benchmarks [76.35271072704384]
  Deep learning models perform poorly when applied to videos with rare scenes or objects. We tackle this problem from two different angles: algorithm and dataset. We show that the debiased representation can generalize better when transferred to other datasets and tasks.
  arXiv  Detail & Related papers  (2022-09-20T00:30:35Z)
• The Two Dimensions of Worst-case Training and the Integrated Effect for Out-of-domain Generalization [95.34898583368154]
  We propose a new, simple yet effective, generalization to train machine learning models. We name our method W2D following the concept of "Worst-case along Two Dimensions"
  arXiv  Detail & Related papers  (2022-04-09T04:14:55Z)
• Perspective: A Phase Diagram for Deep Learning unifying Jamming, Feature Learning and Lazy Training [4.318555434063275]
  Deep learning algorithms are responsible for a technological revolution in a variety of tasks including image recognition or Go playing. Yet, why they work is not understood. Ultimately, they manage to classify data lying in high dimension -- a feat generically impossible. We argue that different learning regimes can be organized into a phase diagram.
  arXiv  Detail & Related papers  (2020-12-30T11:00:36Z)
• Recurrent Multi-view Alignment Network for Unsupervised Surface Registration [79.72086524370819]
  Learning non-rigid registration in an end-to-end manner is challenging due to the inherent high degrees of freedom and the lack of labeled training data. We propose to represent the non-rigid transformation with a point-wise combination of several rigid transformations. We also introduce a differentiable loss function that measures the 3D shape similarity on the projected multi-view 2D depth images.
  arXiv  Detail & Related papers  (2020-11-24T14:22:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.