Theoretical Limitations of Ensembles in the Age of Overparameterization

• URL: http://arxiv.org/abs/2410.16201v1
• Date: Mon, 21 Oct 2024 17:03:20 GMT
• Title: Theoretical Limitations of Ensembles in the Age of Overparameterization
• Authors: Niclas Dern, John P. Cunningham, Geoff Pleiss,
• Abstract summary: Recent empirical studies find that modern ensembles of neural networks may not provide any inherent generalization advantage over single but larger neural networks. We prove that infinite ensembles of over parameterized RF regressors become pointwise equivalent to (single) infinite-width RF regressors. As a consequence, we can characterize the predictive variance amongst ensemble members, and demonstrate that it quantifies the expected effects of increasing capacity.
• Score: 28.965084623872464
• License:
• Abstract: Classic tree-based ensembles generalize better than any single decision tree. In contrast, recent empirical studies find that modern ensembles of (overparameterized) neural networks may not provide any inherent generalization advantage over single but larger neural networks. This paper clarifies how modern overparameterized ensembles differ from their classic underparameterized counterparts, using ensembles of random feature (RF) regressors as a basis for developing theory. In contrast to the underparameterized regime, where ensembling typically induces regularization and increases generalization, we prove that infinite ensembles of overparameterized RF regressors become pointwise equivalent to (single) infinite-width RF regressors. This equivalence, which is exact for ridgeless models and approximate for small ridge penalties, implies that overparameterized ensembles and single large models exhibit nearly identical generalization. As a consequence, we can characterize the predictive variance amongst ensemble members, and demonstrate that it quantifies the expected effects of increasing capacity rather than capturing any conventional notion of uncertainty. Our results challenge common assumptions about the advantages of ensembles in overparameterized settings, prompting a reconsideration of how well intuitions from underparameterized ensembles transfer to deep ensembles and the overparameterized regime.

Related papers

• Revisiting Optimism and Model Complexity in the Wake of Overparameterized Machine Learning [6.278498348219108]
  We revisit model complexity from first principles, by first reinterpreting and then extending the classical statistical concept of (effective) degrees of freedom. We demonstrate the utility of our proposed complexity measures through a mix of conceptual arguments, theory, and experiments.
  arXiv  Detail & Related papers  (2024-10-02T06:09:57Z)
• A Non-negative VAE:the Generalized Gamma Belief Network [49.970917207211556]
  The gamma belief network (GBN) has demonstrated its potential for uncovering multi-layer interpretable latent representations in text data. We introduce the generalized gamma belief network (Generalized GBN) in this paper, which extends the original linear generative model to a more expressive non-linear generative model. We also propose an upward-downward Weibull inference network to approximate the posterior distribution of the latent variables.
  arXiv  Detail & Related papers  (2024-08-06T18:18:37Z)
• Scaling and renormalization in high-dimensional regression [72.59731158970894]
  This paper presents a succinct derivation of the training and generalization performance of a variety of high-dimensional ridge regression models. We provide an introduction and review of recent results on these topics, aimed at readers with backgrounds in physics and deep learning.
  arXiv  Detail & Related papers  (2024-05-01T15:59:00Z)
• PAC-Chernoff Bounds: Understanding Generalization in the Interpolation Regime [6.645111950779666]
  This paper introduces a distribution-dependent PAC-Chernoff bound that exhibits perfect tightness for interpolators. We present a unified theoretical framework revealing why certain interpolators show an exceptional generalization, while others falter.
  arXiv  Detail & Related papers  (2023-06-19T14:07:10Z)
• Instance-Dependent Generalization Bounds via Optimal Transport [51.71650746285469]
  Existing generalization bounds fail to explain crucial factors that drive the generalization of modern neural networks. We derive instance-dependent generalization bounds that depend on the local Lipschitz regularity of the learned prediction function in the data space. We empirically analyze our generalization bounds for neural networks, showing that the bound values are meaningful and capture the effect of popular regularization methods during training.
  arXiv  Detail & Related papers  (2022-11-02T16:39:42Z)
• Provably tuning the ElasticNet across instances [53.0518090093538]
  We consider the problem of tuning the regularization parameters of Ridge regression, LASSO, and the ElasticNet across multiple problem instances. Our results are the first general learning-theoretic guarantees for this important class of problems.
  arXiv  Detail & Related papers  (2022-07-20T21:22:40Z)
• Interpolation can hurt robust generalization even when there is no noise [76.3492338989419]
  We show that avoiding generalization through ridge regularization can significantly improve generalization even in the absence of noise. We prove this phenomenon for the robust risk of both linear regression and classification and hence provide the first theoretical result on robust overfitting.
  arXiv  Detail & Related papers  (2021-08-05T23:04:15Z)
• Nonasymptotic theory for two-layer neural networks: Beyond the bias-variance trade-off [10.182922771556742]
  We present a nonasymptotic generalization theory for two-layer neural networks with ReLU activation function. We show that overparametrized random feature models suffer from the curse of dimensionality and thus are suboptimal.
  arXiv  Detail & Related papers  (2021-06-09T03:52:18Z)
• Post-mortem on a deep learning contest: a Simpson's paradox and the complementary roles of scale metrics versus shape metrics [61.49826776409194]
  We analyze a corpus of models made publicly-available for a contest to predict the generalization accuracy of neural network (NN) models. We identify what amounts to a Simpson's paradox: where "scale" metrics perform well overall but perform poorly on sub partitions of the data. We present two novel shape metrics, one data-independent, and the other data-dependent, which can predict trends in the test accuracy of a series of NNs.
  arXiv  Detail & Related papers  (2021-06-01T19:19:49Z)
• Understanding Double Descent Requires a Fine-Grained Bias-Variance Decomposition [34.235007566913396]
  We describe an interpretable, symmetric decomposition of the variance into terms associated with the labels. We find that the bias decreases monotonically with the network width, but the variance terms exhibit non-monotonic behavior. We also analyze the strikingly rich phenomenology that arises.
  arXiv  Detail & Related papers  (2020-11-04T21:04:02Z)
• Asymptotics of Ridge (less) Regression under General Source Condition [26.618200633139256]
  We consider the role played by the structure of the true regression parameter. We show that (no regularisation) can be optimal even with bounded signal-to-noise ratio (SNR) This contrasts with previous work considering ridge regression with isotropic prior, in which case is only optimal in the limit of infinite SNR.
  arXiv  Detail & Related papers  (2020-06-11T13:00:21Z)

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.