Proposes a new calibration error estimator for deep neural networks.
problem Improves calibration of deep neural networks, especially for canonical calibration.
method Uses a Dirichlet kernel density estimate to create a low-bias, trainable calibration error estimator.
result Asymptotically converges to true Lp calibration error, enabling efficient estimation and mini-batch updates. This paper rethinks confidence calibration under covariate shifts.
problem Calibration methods struggle with covariate shifts and unstable importance weighting.
method Derives Expectation consistency condition and proposes Expectation consistency loss (ECL).
result ECL loss is compatible with various types of calibration and has the same sample complexity as ECE.
Proposes h-calibration for improving miscalibrated probability outputs of neural networks.
problem Improving reliability of probability outputs from neural networks.
method Probabilistic learning framework for calibration, including a simple yet effective post-hoc algorithm.
result Significantly better performance than traditional methods, validated by experiments.
The paper explores connections between quaternionic and Cayley calibrations in dimensions 8 and 16.
problem Exploring connections between quaternionic and Cayley calibrations in dimensions 8 and 16.
method Starting from collections of 'Kähler 2-forms', the paper constructs canonical 4-forms and calibrated 4-planes in dimensions 8 and 16.
result Explicit formulas for canonical 4-forms ΦSpin(8) and ΦSpin(7)U(1) are derived, and their calibrated 4-planes are characterized. The squashed 7-sphere S7 is a 7-sphere with an Einstein metric given by the canonical variation and its cone R8−{0} has full holonomy Spin(7). There is a canonical calibrating 4-form Φ on R8−{0}. A minimal 3-submanifold in S7 is called associative if its cone …
Optimizes calibration error estimators for better classifier trustworthiness.
problem Lack of guidance on selecting and tuning calibration error estimators.
method Reformulates calibration estimation as a regression problem with i.i.d. input pairs.
result Demonstrates the effectiveness of optimized calibration estimators on image classification tasks.
This work proves L2-regularized ERM controls smCE without post-hoc correction.
problem Calibration of predicted probabilities in machine learning models.
method Canonical L2-regularized empirical risk minimization. result Theoretical proof that smCE is controlled by ERM without post-hoc correction.
Study calibrated geometry in hyperkähler cones and their related spaces.
problem Characterize submanifolds in hyperkähler cones and related spaces.
method Systematic study of calibrated geometry in hyperkähler cones, 3-Sasakian manifolds, and twistor spaces.
result Obtain new characterizations of complex Lagrangian and complex isotropic cones in hyperkähler cones.
It is well known that there is a unique Spin(9)-invariant 8-form on the octonionic plane that naturally yields a canonical differential 8-form on any Riemannian manifold with a weak Spin(9)-structure. Over the decades, this invariant has been studied extensively and described in several equivalent ways. In the pres…
We show the total space of the canonical line bundle L of a Kahler-Einstein manifold Xn supports integrable SU(n+1) structures, or Calabi-Yau structures. The canonical real line bundle L⊂L over a minimal Lagrangian submanifold M⊂X is calibrated in this setting and hence can …
We study massless deformations of generalized calibrated cycles, which describe, in the language of generalized complex geometry, supersymmetric D-branes in N=1 supersymmetric compactifications with fluxes. We find that the deformations are classified by the first cohomology group of a Lie algebroid canonically associa…
In this paper we study the continuum time dynamics of a stock in a market where agents behavior is modeled by a Minority Game and a Grand Canonical Minority Game. The dynamics derived is a generalized geometric Brownian motion; from the Black & Scholes formula the calibration of both the Minority Game and the Grand Can…
The paper constructs a lamination related to minimal hypersurfaces calibrated by a cohomology class.
problem Understanding the geometry of stable norm balls constrained by manifold topology.
method Constructing a lamination λρ of minimal hypersurfaces calibrated by ρ. result Establishes a close analogy between stable norm and earthquake norms.
The paper studies affine models driven by independent Lévy processes and their calibration.
problem Characterizing and classifying affine models driven by Lévy processes.
method Analyzing the short rate equation with independent Lévy processes and characterizing the generator.
result A precise form of the generator and classification of affine models with canonical representations.
The paper studies p-harmonic functions and their conjugates, showing they converge to calibrations of laminations.
problem Behavior of q-harmonic functions and their conjugates in the limit as qo1. method Analysis of p-harmonic conjugates and their convergence to calibrations of laminations. result The laminations calibrated by the limiting p-harmonic conjugates are exactly those arising from the 1-Laplacian. Let H be the hyperbolic space of dimension n+1. A geodesic foliation of H is given by a smooth unit vector field on H all of whose integral curves are geodesics. Each geodesic foliation of H determines an n-dimensional submanifold M of the 2n-dimensional manifold L of all the oriented geodesics of H (up to orientation …
Calibrated models can lead to miscalibrated aggregations in strategic interactions.
problem Miscalibration in aggregated predictions from multiple calibrated models.
method Analysis of strategic interactions between calibrated predictors, proving conditions for miscalibration and comparing VCG and Brier-score aggregation methods.
result VCG aggregation method outperforms Brier-score in strategic settings, providing robustness and comparable accuracy.
We show that the properties of Lagrangian mean curvature flow are a special case of a more general phenomenon, concerning couplings between geometric flows of the ambient space and of totally real submanifolds. Both flows are driven by ambient Ricci curvature or, in the non-Kähler case, by its analogues. To this end we…
We present a construction of a canonical G_2 structure on the unit sphere tangent bundle S_M of any given orientable Riemannian 4-manifold M. Such structure is never geometric or 1-flat, but seems full of other possibilities. We start by the study of the most basic properties of our construction. The structure is co-ca…
D-GCCA improves multi-view data analysis by separating common and distinctive components.
problem Analyzing multi-view high-dimensional data with latent factors.
method Decomposes each view's data matrix into common and distinctive sources with orthogonality constraints.
result Consistent estimators with good performance and efficient computation.
In this paper we will investigate torus actions on complete manifolds with calibrations. For Calabi-Yau manifolds M^2n with a Hamiltonian structure-preserving k-torus action we show that any symplectic reduction has a natural holomorphic volume form. Moreover Special Lagrangian (SLag) submanifolds of the reduction lift…
A hypercomplex manifold is a manifold equipped with a triple of complex structures satisfying the quaternionic relations. A holomorphic Lagrangian variety on a hypercomplex manifold with trivial canonical bundle is a holomorphic subvariety which is calibrated by a form associated with the holomorphic volume form; this …
Plots show miscalibration directly as slopes of secant lines.
problem Detecting discrepancies between probabilistic predictions and actual outcomes.
method Cumulative differences between observed and expected values displayed as slopes of secant lines.
result Directly shows miscalibration without binning or kernel density estimation.
Proposes new method for calibrating treatment effect predictors.
problem Calibrating predictors of heterogeneous treatment effects.
method Causal isotonic calibration and cross-calibration.
result Achieves fast calibration rates under weak conditions.
We derive a forward equation for arbitrage-free barrier option prices, in terms of Markovian projections of the stochastic volatility process, in continuous semi-martingale models. This provides a Dupire-type formula for the coefficient derived by Brunick and Shreve for their mimicking diffusion and can be interpreted …
Proposes top-label calibration and M2B framework for multiclass to binary calibration.
problem Multiclass calibration and interpretation issues.
method Top-label calibration and M2B reduction framework.
result M2B + HB achieves lower calibration error than other methods.
Study explores calibration properties in neural architectures.
problem Calibration issues in deep neural networks despite improved accuracy.
method Leverages Neural Architecture Search (NAS) to evaluate 117,702 neural networks.
result Identifies key architectural designs beneficial for calibration.
A new perfectly truthful calibration measure improves prediction reliability.
problem Improving the reliability of predictions by ensuring they are conditionally unbiased.
method Designing a simple, perfectly truthful calibration measure called ATB.
result ATB is the first perfectly truthful calibration measure in the batch setting.
New truthful calibration errors improve model ranking in multiclass prediction.
problem Non-truthful calibration errors can mislead model comparisons.
method Introduced perfectly truthful calibration errors for multiclass predictions.
result Truthful calibration errors preserve decision-theoretic dominance and stabilize model rankings.
New framework for evaluating multiclass classifier calibration.
problem Ensuring classifiers are well-calibrated for trustworthy predictions.
method Utility Calibration framework that measures calibration error relative to a utility function.
result Unified and robust interpretation of existing calibration metrics.
We propose a new framework to improve the calibration of neural networks.
problem Improving the accuracy of model confidence predictions.
method Introducing a differentiable surrogate for expected calibration error (DECE) and a meta-learning framework to optimise model hyper-parameters for validation set calibration.
result Achieved competitive performance with existing calibration approaches.
Certified calibration methods protect model confidence from adversarial attacks.
problem Adversarial attacks degrade model calibration, reducing confidence in predictions.
method Developed certified calibration methods to provide worst-case bounds on calibration under adversarial perturbations.
result Certified calibration methods produce analytic and approximate bounds for the Brier score and expected calibration error.
Meta-Cal improves post-hoc calibration of neural networks.
problem Improving the accuracy of uncalibrated neural network predictions.
method Meta-Cal uses a base calibrator and a ranking model with constraints to provide high-probability bounds.
result Meta-Cal significantly outperforms existing methods in post-hoc multi-class classification calibration.
Dead-Direction Signatures (DDS) provide a cheap, closed-form spectral reading of a network's singular complexity.
problem Estimating the complexity of deep networks through their loss singularities.
method DDS replaces the SGLD posterior chain with spectral linear algebra.
result DDS observables rank-track the network's singular complexity at the framework-predicted sign.
Post-processing predictors reduces calibration errors for decision-making.
problem Predictors with low calibration error for machine learning may have high error for decision-making.
method Post-processing with ε distance to calibration adds noise to make predictions differentially private.
result Post-processing achieves O(√ε) ECE and CDL, asymptotically optimal.
A new calibration metric bridges testability and actionability.
problem Combining testability and actionable insights for forecast probabilities.
method Cutoff Calibration Error (CCE) that assesses calibration over intervals of forecasted probabilities.
result Cutoff Calibration Error is both testable and actionable.
New study on neural network calibration, linking it to generalization gap.
problem Neural networks lack strong guarantees on calibration.
method Decomposed calibration error into train set and generalization gap.
result Models with small generalization gap are well-calibrated.
A new method for multiclass calibration using vector quantization.
problem Challenges in multiclass calibration, especially in high-stakes settings.
method Compositional approach via Vector Quantization (VQ) to learn region-specific calibration maps.
result Significant improvements in local calibration with competitive global calibration and predictive performance.
This post introduces model calibration and evaluation measures, highlighting issues with a common measure.
problem Ensuring model confidence accurately reflects true outcomes.
method Explains common calibration definition, ECE, and its drawbacks.
result New evaluation measures needed for comprehensive model calibration.
Unified calibration metrics improve forecast sharpness and accuracy.
problem Improving the sharpness of probabilistic forecasts while maintaining calibration.
method Kernel-based calibration metrics that unify and generalize existing methods for classification and regression.
result Enhanced calibration, sharpness, and decision-making across various tasks.
New decision-theoretic calibration error metric improves prediction reliability.
problem Improving the reliability of predictions for decision-making.
method Proposed Calibration Decision Loss (CDL) and an efficient algorithm to achieve near-optimal CDL.
result Near-optimal CDL guarantees vanishing payoff loss from miscalibration.
Cone structures over minimal products can't be calibrated smoothly.
problem Calibrating cones over minimal products with smooth calibrations.
method Extending a key result from [Zha26], showing obstruction.
result Cone structures over minimal products cannot be calibrated by smooth calibrations.
Post-hoc calibration improves uncertainty under domain shift.
problem Improving uncertainty calibration under domain shift.
method Apply perturbations to validation set before post-hoc calibration.
result Perturbation step results in better calibration under domain shift.
Smooth calibration improves forecast reliability even with leaked information.
problem Improving forecast reliability with leaked information.
method Combining nearby forecasts to ensure smooth calibration, which can be guaranteed by deterministic procedures.
result Smooth calibration can be guaranteed by deterministic procedures even with leaked forecasts, and it yields uncoupled finite-memory dynamics in games.
Planes are the only calibrated submanifolds with flat normal bundles.
problem Characterizing submanifolds with specific geometric properties.
method Using constant-coefficient differential forms and parallel calibrations.
result Calibrated submanifolds with flat normal bundles are planes.
We describe a family of calibrations arising naturally on a hyperkähler manifold M. These calibrations calibrate the holomorphic Lagrangian, holomorphic isotropic and holomorphic coisotropic subvarieties. When M is an HKT (hyperkaehler with torsion) manifold with holonomy SL(n,H), we construct another fam…
Simple proof shows forecasts can be calibrated in a few periods.
problem Ensuring forecasts are calibrated over multiple periods.
method Uses minimax theorem to prove existence and guarantees calibration error.
result Calibration can be achieved in N3 periods with error at most 1/N. Survey of methods to calibrate neural network predictions.
problem Ensuring neural networks provide accurate confidence levels.
method Empirical comparison of calibration methods.
result Various techniques for calibrating neural networks.