Feature Understanding and Sparsity Enhancement via 2-Layered kernel machines (2L-FUSE)

• URL: http://arxiv.org/abs/2509.07806v1
• Date: Tue, 09 Sep 2025 14:45:10 GMT
• Title: Feature Understanding and Sparsity Enhancement via 2-Layered kernel machines (2L-FUSE)
• Abstract summary: We propose a novel sparsity enhancement strategy for regression tasks, based on learning a data-adaptive kernel metric.<n>The resulting shape matrix defines a Mahalanobis-type deformation of the input space, which is factorized via an eigen-decomposition.<n>Experiments on synthetic and applications to real datasets of geomagnetic storms demonstrate that our approach achieves minimal yet highly informative feature sets without losing predictive performance.
• Score: 3.674863913115431
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We propose a novel sparsity enhancement strategy for regression tasks, based on learning a data-adaptive kernel metric, i.e., a shape matrix, through 2-Layered kernel machines. The resulting shape matrix, which defines a Mahalanobis-type deformation of the input space, is then factorized via an eigen-decomposition, allowing us to identify the most informative directions in the space of features. This data-driven approach provides a flexible, interpretable and accurate feature reduction scheme. Numerical experiments on synthetic and applications to real datasets of geomagnetic storms demonstrate that our approach achieves minimal yet highly informative feature sets without losing predictive performance.

Related papers

• How does feature learning reshape the function space? [17.342357780996174]
  We study how the function space spanned by the features of a two-layer neural network evolves during gradient descent training.<n>Our results provide a precise function-space perspective on early-stage feature learning.
  arXiv  Detail & Related papers  (2026-05-18T00:41:20Z)
• A Mathematical Optimization Approach to Multisphere Support Vector Data Description [1.9499277906326784]
  We provide a primal formulation, in the shape of a Mixed Second Order Cone model, that constructs Euclidean hyperspheres to identify anomalous observations.<n>We develop a dual model that enables the application of the kernel trick, thus allowing for the detection of outliers within complex, non-linear data structures.
  arXiv  Detail & Related papers  (2025-07-15T08:57:27Z)
• Localized Gaussians as Self-Attention Weights for Point Clouds Correspondence [92.07601770031236]
  We investigate semantically meaningful patterns in the attention heads of an encoder-only Transformer architecture.<n>We find that fixing the attention weights not only accelerates the training process but also enhances the stability of the optimization.
  arXiv  Detail & Related papers  (2024-09-20T07:41:47Z)
• Optimization of a Hydrodynamic Computational Reservoir through Evolution [58.720142291102135]
  We interface with a model of a hydrodynamic system, under development by a startup, as a computational reservoir. We optimized the readout times and how inputs are mapped to the wave amplitude or frequency using an evolutionary search algorithm. Applying evolutionary methods to this reservoir system substantially improved separability on an XNOR task, in comparison to implementations with hand-selected parameters.
  arXiv  Detail & Related papers  (2023-04-20T19:15:02Z)
• Multi-Linear Kernel Regression and Imputation in Data Manifolds [12.15802365851407]
  This paper introduces an efficient multi-linear nonparametric approximation framework for data regression and imputation, and its application to dynamic magnetic-resonance imaging (dMRI) Data features are assumed to reside in or close to a smooth manifold embedded in a kernel reproducing Hilbert space. Landmark points are identified to describe the point cloud of features by linear approximating patches which mimic the concept of tangent spaces to smooth. The multi-linear model effects dimensionality reduction, enables efficient computations, and extracts data patterns and their geometry without any training data or additional information.
  arXiv  Detail & Related papers  (2023-04-06T12:58:52Z)
• MFAI: A Scalable Bayesian Matrix Factorization Approach to Leveraging Auxiliary Information [8.42894516984735]
  We propose to integrate gradient boosted trees in the probabilistic matrix factorization framework to leverage auxiliary information (MFAI) MFAI naturally inherits several salient features of gradient boosted trees, such as the capability of flexibly modeling nonlinear relationships. MFAI is computationally efficient and scalable to large datasets by exploiting variational inference.
  arXiv  Detail & Related papers  (2023-03-05T03:26:14Z)
• Fundamental Limits of Two-layer Autoencoders, and Achieving Them with Gradient Methods [91.54785981649228]
  This paper focuses on non-linear two-layer autoencoders trained in the challenging proportional regime. Our results characterize the minimizers of the population risk, and show that such minimizers are achieved by gradient methods. For the special case of a sign activation function, our analysis establishes the fundamental limits for the lossy compression of Gaussian sources via (shallow) autoencoders.
  arXiv  Detail & Related papers  (2022-12-27T12:37:34Z)
• Accelerated structured matrix factorization [0.0]
  Matrix factorization exploits the idea that, in complex high-dimensional data, the actual signal typically lies in lower-dimensional structures. By exploiting Bayesian shrinkage priors, we devise a computationally convenient approach for high-dimensional matrix factorization. The dependence between row and column entities is modeled by inducing flexible sparse patterns within factors.
  arXiv  Detail & Related papers  (2022-12-13T11:35:01Z)
• Towards Open-World Feature Extrapolation: An Inductive Graph Learning Approach [80.8446673089281]
  We propose a new learning paradigm with graph representation and learning. Our framework contains two modules: 1) a backbone network (e.g., feedforward neural nets) as a lower model takes features as input and outputs predicted labels; 2) a graph neural network as an upper model learns to extrapolate embeddings for new features via message passing over a feature-data graph built from observed data.
  arXiv  Detail & Related papers  (2021-10-09T09:02:45Z)
• Feature Weighted Non-negative Matrix Factorization [92.45013716097753]
  We propose the Feature weighted Non-negative Matrix Factorization (FNMF) in this paper. FNMF learns the weights of features adaptively according to their importances. It can be solved efficiently with the suggested optimization algorithm.
  arXiv  Detail & Related papers  (2021-03-24T21:17:17Z)
• Two-Dimensional Semi-Nonnegative Matrix Factorization for Clustering [50.43424130281065]
  We propose a new Semi-Nonnegative Matrix Factorization method for 2-dimensional (2D) data, named TS-NMF. It overcomes the drawback of existing methods that seriously damage the spatial information of the data by converting 2D data to vectors in a preprocessing step.
  arXiv  Detail & Related papers  (2020-05-19T05:54:14Z)

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.