Covariance Last-Layer Ensembles: Function-Space Diversity for Efficient Uncertainty Quantification

• URL: http://arxiv.org/abs/2607.23856v2
• Date: Tue, 28 Jul 2026 15:09:18 GMT
• Title: Covariance Last-Layer Ensembles: Function-Space Diversity for Efficient Uncertainty Quantification
• Abstract summary: Cov-LLE restores function-space diversity that weight-orthonormality cannot, and at matched $K$ recovers much of the diversity and calibration of a deep ensemble.<n> Viewing OC as a last-layer ensemble also organizes detectors into a two-axis taxonomy.
• Score: 0.8803840644459077
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: A Last-Layer Ensemble (LLE), $K$ linear units on one shared frozen feature map, is an efficient single-pass approach to the disagreement-based epistemic uncertainty for out-of-distribution (OOD) detection. Its weakness is that members share the backbone gradient and can converge toward the same function, collapsing the inter-member diversity the signal depends on. Whether last-layer diversity can be restored, and what mitigates the collapse, is an open question. The weight-orthonormality defining Orthonormal Certificates (OC), the weight-orthonormal special case of the LLE, is only an indirect correction; it decorrelates the weights of the members, not their predictions. Here, we instead target the collapse directly in function space, with a Covariance Last-Layer Ensemble (cov-LLE) that places a direct covariance penalty on member activations. Cov-LLE restores the function-space diversity that weight-orthonormality cannot, and at matched $K$ recovers much of the diversity and calibration of a deep ensemble at $1\times$ backbone cost (in-distribution prediction variance $0.05\!\to\!9.3$ vs. $22.1$ ($\times10^{-3}$), and ECE $0.135\!\to\!0.090$ vs. $0.035$, for a $K\times$-cost deep ensemble), at no cost to accuracy. Viewing OC as a last-layer ensemble also organizes detectors into a two-axis taxonomy (by how their units are trained and how their outputs are scored) and exposes the OC score as a magnitude, motivating a scale-invariant, label-free direction score that repairs its near-OOD failure, adding $+0.16$ to $+0.18$ ROC AUC on every backbone.

Related papers

• Sharp Rates and a One-Line Correction for Spectral Representation Learning [8.921166277011347]
  Self-supervised encoder is trained once, frozen, and reused through lightweight probes on tasks nobody named at training time.<n> Canonical correlation analysis, HGR maximal correlation, and the population optimum of the spectral contrastive loss all return the top-$k$ singular subspace of a cross-view dependence operator.<n>We show isotropy is the wrong hypothesis. The prior enters the transfer risk only through the task covariance $=mathbbE[top]$, and only through its compression onto the operator's leading singular directions.
  arXiv  Detail & Related papers  (2026-09-14T16:26:55Z)
• How Wrong Can a Good Predictor Be? Diverging Updates with Vanishing Predictive KL [8.929322780616545]
  An unbounded gap between the update maps can coexist with vanishing predictive KL for every fixed finite $Kge2$ in a stationary symmetric Gaussian HMM.<n>The construction is fixed in $K$ and does not provide a universal criterion for when compression is harmless or characterize when internal gaps must incur task loss.
  arXiv  Detail & Related papers  (2026-09-10T06:22:53Z)
• Between-User Collapse Under Popularity-Biased Feedback: A Centered-Covariance Theorem and Computable Phase Boundary [0.0]
  We study how popularity-biased BPR training reshapes the between-user geometry of collaborative-filtering embeddings.<n>We prove that under popularity-biased feedback with stationary items, $C$ converges to a steady state.<n>We then examine the limits of the effect.
  arXiv  Detail & Related papers  (2026-08-03T17:33:25Z)
• Round-Trip Consistency: Bidirectional Diffusion Models Can Predict Their Own Rollout Errors [51.56484100374058]
  Autoregressive models accumulate error over long rollouts, yet at deployment there is no ground truth to measure it against.<n>We train a single conditional latent diffusion model that steps a dynamical system forward or backward in time via a direction flag.<n>We show that this bidirectionality supplies a measurement-free test-time error signal.
  arXiv  Detail & Related papers  (2026-08-01T13:49:46Z)
• Saturation Makes Quantization Error Additive: A Coverage Model with a Certificate [0.0]
  Mixed-precision quantization must decide which parts of a model to keep at higher precision.<n>We show that the loss from quantizing a set of layers can be reconstructed from per-layer or pairwise sensitivities measured in isolation.<n>We propose the coverage model $f(S)=cbigl (1-prod_iin S (1-a_i)bigr)$, which reproduces the measured variance profile of $f$ to within a few percent from its $L$ fitted break-rates.
  arXiv  Detail & Related papers  (2026-07-14T02:08:33Z)
• A Spectral Phase Diagram for Binary Few-Shot Classification: Intrinsic Dimensionality, Geometric Saturation, and Representational Diagnosis [0.0]
  The saturation index $S(K) = operatornameerank(widehat_W(K)) / K$ measures the ratio of the effective rank of the pooled within-class sample covariance to the shot count.<n>The index is computable in $O(d3)$ time from support features alone, requiring no test labels or trained classifier.
  arXiv  Detail & Related papers  (2026-06-12T16:46:24Z)
• Is Spurious Correlation Removal Always Learnable? [56.28155520961125]
  Invariant learning can fail even when the structure is statistically identifiable.<n>We show a conditional computational barrier: under a black-box samplable supervised sparse recovery primitive, there exist emphamplable multi-dimensional environments.<n>Synthetic and real datasets illustrate the predicted gaps and transitions and motivate simple diversity diagnostics.
  arXiv  Detail & Related papers  (2026-06-11T05:49:43Z)
• Hidden-State Privacy Has an Empty Middle [51.56484100374058]
  We prove a complementary Fisher-ball lower bound: every full-rank Gaussian release at $O(1)$ Fisher utility admits a direction whose Mahalanobis signal grows linearly in hidden width.<n>A split-memory transformer trained from scratch reaches $G_mathrmMah in [20, 33]$ at 90M and maintains a $6$--$24times$ advantage over same-budget GPT baselines from 30M to 1B at a fixed-token language loss penalty.
  arXiv  Detail & Related papers  (2026-05-21T20:12:09Z)
• Correction and Corruption: A Two-Rate View of Error Flow in LLM Protocols [51.56484100374058]
  We propose a paired-outcome measurement interface for auditing a single protocol step on exact-match tasks.<n>For each instance, the interface records a baseline correctness bit and a post-step correctness bit.<n>These rates predict accuracy changes and define a reusable empirical interface testable across seeds, mixtures, and pipelines.
  arXiv  Detail & Related papers  (2026-04-20T13:25:40Z)
• Tight Convergence Rates for Online Distributed Linear Estimation with Adversarial Measurements [66.94250413799232]
  We study mean estimation of a random vector $X$ in a distributed parameter-server-worker setup.<n>Key challenges are adversarial measurements and asynchrony.<n>Results provide a unified finite-time characterization of robustness, identifiability, and statistical efficiency in distributed linear estimation with adversarial workers.
  arXiv  Detail & Related papers  (2026-04-07T11:45:55Z)
• When Does Margin Clamping Affect Training Variance? Dataset-Dependent Effects in Contrastive Forward-Forward Learning [0.0]
  Contrastive Forward-Forward (CFF) learning trains Vision Transformers layer by layer against supervised contrastive objectives.<n>We focus on one implementation detail: the positive-pair margin in the contrastive loss is applied through saturating similarity clamping.<n>We prove that an alternative formulation, subtracting the margin after the log-probability, is gradient-neutral under the mean-over-positives reduction.
  arXiv  Detail & Related papers  (2026-03-01T07:00:38Z)
• Wasserstein-regularized Conformal Prediction under General Distribution Shift [21.29977879431466]
  Conformal prediction yields a prediction set with guaranteed $1-alpha$ coverage of the true target under the i.i.d. assumption.<n>We propose a Wasserstein distance-based upper bound of the coverage gap and analyze the bound using probability measure pushforwards.<n>We exploit the separation to design an algorithm based on importance weighting and regularized representation learning (WR-CP) to reduce the Wasserstein bound with a finite-sample error bound.
  arXiv  Detail & Related papers  (2025-01-23T07:29:44Z)
• Improved Algorithm for Adversarial Linear Mixture MDPs with Bandit Feedback and Unknown Transition [71.33787410075577]
  We study reinforcement learning with linear function approximation, unknown transition, and adversarial losses. We propose a new algorithm that attains an $widetildeO(dsqrtHS3K + sqrtHSAK)$ regret with high probability.
  arXiv  Detail & Related papers  (2024-03-07T15:03:50Z)
• TIC-TAC: A Framework for Improved Covariance Estimation in Deep Heteroscedastic Regression [109.69084997173196]
  Deepscedastic regression involves jointly optimizing the mean and covariance of the predicted distribution using the negative log-likelihood. Recent works show that this may result in sub-optimal convergence due to the challenges associated with covariance estimation. We study two questions: (1) Does the predicted covariance truly capture the randomness of the predicted mean? Our results show that not only does TIC accurately learn the covariance, it additionally facilitates an improved convergence of the negative log-likelihood.
  arXiv  Detail & Related papers  (2023-10-29T09:54:03Z)

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.