Random Features Outperform Linear Models: Effect of Strong Input-Label Correlation in Spiked Covariance Data

• URL: http://arxiv.org/abs/2409.20250v1
• Date: Mon, 30 Sep 2024 12:40:45 GMT
• Title: Random Features Outperform Linear Models: Effect of Strong Input-Label Correlation in Spiked Covariance Data
• Authors: Samet Demir, Zafer Dogan,
• Abstract summary: We show that a high correlation between inputs and labels is a critical factor enabling the RFM to outperform linear models. We show that the RFM performs equivalent to noisy models, where the degree depends on the strength of the correlation between inputs and labels.
• Score: 0.8287206589886879
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Random Feature Model (RFM) with a nonlinear activation function is instrumental in understanding training and generalization performance in high-dimensional learning. While existing research has established an asymptotic equivalence in performance between the RFM and noisy linear models under isotropic data assumptions, empirical observations indicate that the RFM frequently surpasses linear models in practical applications. To address this gap, we ask, "When and how does the RFM outperform linear models?" In practice, inputs often have additional structures that significantly influence learning. Therefore, we explore the RFM under anisotropic input data characterized by spiked covariance in the proportional asymptotic limit, where dimensions diverge jointly while maintaining finite ratios. Our analysis reveals that a high correlation between inputs and labels is a critical factor enabling the RFM to outperform linear models. Moreover, we show that the RFM performs equivalent to noisy polynomial models, where the polynomial degree depends on the strength of the correlation between inputs and labels. Our numerical simulations validate these theoretical insights, confirming the performance-wise superiority of RFM in scenarios characterized by strong input-label correlation.

Related papers

• Learning Divergence Fields for Shift-Robust Graph Representations [73.11818515795761]
  In this work, we propose a geometric diffusion model with learnable divergence fields for the challenging problem with interdependent data. We derive a new learning objective through causal inference, which can guide the model to learn generalizable patterns of interdependence that are insensitive across domains.
  arXiv  Detail & Related papers  (2024-06-07T14:29:21Z)
• Deep Latent Force Models: ODE-based Process Convolutions for Bayesian Deep Learning [0.0]
  The deep latent force model (DLFM) is a deep Gaussian process with physics-informed kernels at each layer. We present empirical evidence of the capability of the DLFM to capture the dynamics present in highly nonlinear real-world time series data. We find that the DLFM is capable of achieving comparable performance to a range of non-physics-informed probabilistic models.
  arXiv  Detail & Related papers  (2023-11-24T19:55:57Z)
• A PAC-Bayesian Perspective on the Interpolating Information Criterion [54.548058449535155]
  We show how a PAC-Bayes bound is obtained for a general class of models, characterizing factors which influence performance in the interpolating regime. We quantify how the test error for overparameterized models achieving effectively zero training error depends on the quality of the implicit regularization imposed by e.g. the combination of model, parameter-initialization scheme.
  arXiv  Detail & Related papers  (2023-11-13T01:48:08Z)
• Accuracy on the Curve: On the Nonlinear Correlation of ML Performance Between Data Subpopulations [24.579430688134185]
  We show that correlation between in-distribution (ID) and out-of-distribution (OOD) accuracies is more nuanced under subpopulation shifts. Our work highlights the importance of understanding the nonlinear effects of model improvement on performance in different subpopulations.
  arXiv  Detail & Related papers  (2023-05-04T17:00:17Z)
• Biases in Inverse Ising Estimates of Near-Critical Behaviour [0.0]
  Inverse inference allows pairwise interactions to be reconstructed from empirical correlations. We show that estimators used for this inference, such as Pseudo-likelihood (PLM), are biased. Data-driven methods are explored and applied to a functional magnetic resonance imaging (fMRI) dataset from neuroscience.
  arXiv  Detail & Related papers  (2023-01-13T14:01:43Z)
• A Bayesian Framework on Asymmetric Mixture of Factor Analyser [0.0]
  This paper introduces an MFA model with a rich and flexible class of skew normal (unrestricted) generalized hyperbolic (called SUNGH) distributions. The SUNGH family provides considerable flexibility to model skewness in different directions as well as allowing for heavy tailed data. Considering factor analysis models, the SUNGH family also allows for skewness and heavy tails for both the error component and factor scores.
  arXiv  Detail & Related papers  (2022-11-01T20:19:52Z)
• Efficient CDF Approximations for Normalizing Flows [64.60846767084877]
  We build upon the diffeomorphic properties of normalizing flows to estimate the cumulative distribution function (CDF) over a closed region. Our experiments on popular flow architectures and UCI datasets show a marked improvement in sample efficiency as compared to traditional estimators.
  arXiv  Detail & Related papers  (2022-02-23T06:11:49Z)
• Optimal regularizations for data generation with probabilistic graphical models [0.0]
  Empirically, well-chosen regularization schemes dramatically improve the quality of the inferred models. We consider the particular case of L 2 and L 1 regularizations in the Maximum A Posteriori (MAP) inference of generative pairwise graphical models.
  arXiv  Detail & Related papers  (2021-12-02T14:45:16Z)
• Accuracy on the Line: On the Strong Correlation Between Out-of-Distribution and In-Distribution Generalization [89.73665256847858]
  We show that out-of-distribution performance is strongly correlated with in-distribution performance for a wide range of models and distribution shifts. Specifically, we demonstrate strong correlations between in-distribution and out-of-distribution performance on variants of CIFAR-10 & ImageNet. We also investigate cases where the correlation is weaker, for instance some synthetic distribution shifts from CIFAR-10-C and the tissue classification dataset Camelyon17-WILDS.
  arXiv  Detail & Related papers  (2021-07-09T19:48:23Z)
• Quaternion Factorization Machines: A Lightweight Solution to Intricate Feature Interaction Modelling [76.89779231460193]
  factorization machine (FM) is capable of automatically learning high-order interactions among features to make predictions without the need for manual feature engineering. We propose the quaternion factorization machine (QFM) and quaternion neural factorization machine (QNFM) for sparse predictive analytics.
  arXiv  Detail & Related papers  (2021-04-05T00:02:36Z)
• Multiplicative noise and heavy tails in stochastic optimization [62.993432503309485]
  empirical optimization is central to modern machine learning, but its role in its success is still unclear. We show that it commonly arises in parameters of discrete multiplicative noise due to variance. A detailed analysis is conducted in which we describe on key factors, including recent step size, and data, all exhibit similar results on state-of-the-art neural network models.
  arXiv  Detail & Related papers  (2020-06-11T09:58:01Z)

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.