The Sample Complexity of Multicalibration

• URL: http://arxiv.org/abs/2604.21923v1
• Date: Thu, 23 Apr 2026 17:59:01 GMT
• Title: The Sample Complexity of Multicalibration
• Abstract summary: We study the minimax sample complexity of multicalibration in the batch setting.<n>A learner observes $n$ i.i.d. samples from an unknown distribution and must output a predictor whose population multicalibration error, measured by Expected Error (ECE), is at most $varepsilon$ with respect to a given family of groups.<n>We establish matching upper and lower bounds, up to polylogarithmic factors, for a weighted $L_p$ multicalibration metric for all $1 le p le 2$, with optimal $3/p$.
• Score: 9.852468478532652
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We study the minimax sample complexity of multicalibration in the batch setting. A learner observes $n$ i.i.d. samples from an unknown distribution and must output a (possibly randomized) predictor whose population multicalibration error, measured by Expected Calibration Error (ECE), is at most $\varepsilon$ with respect to a given family of groups. For every fixed $κ> 0$, in the regime $|G|\le \varepsilon^{-κ}$, we prove that $\widetildeΘ(\varepsilon^{-3})$ samples are necessary and sufficient, up to polylogarithmic factors. The lower bound holds even for randomized predictors, and the upper bound is realized by a randomized predictor obtained via an online-to-batch reduction. This separates the sample complexity of multicalibration from that of marginal calibration, which scales as $\widetildeΘ(\varepsilon^{-2})$, and shows that mean-ECE multicalibration is as difficult in the batch setting as it is in the online setting, in contrast to marginal calibration which is strictly more difficult in the online setting. In contrast we observe that for $κ= 0$, the sample complexity of multicalibration remains $\widetildeΘ(\varepsilon^{-2})$ exhibiting a sharp threshold phenomenon. More generally, we establish matching upper and lower bounds, up to polylogarithmic factors, for a weighted $L_p$ multicalibration metric for all $1 \le p \le 2$, with optimal exponent $3/p$. We also extend the lower-bound template to a regular class of elicitable properties, and combine it with the online upper bounds of Hu et al. (2025) to obtain matching bounds for calibrating properties including expectiles and bounded-density quantiles.

Related papers

• Optimal Low-Rank Quantum State Tomography with Bounded-Sample Joint Measurements [6.929688137243709]
  We determine the optimal sample complexity of low-rank quantum state tomography when each measurement may act jointly on at most $t$ samples.<n>For the lower bound, we vary the support of a state with fixed uniform spectrum and bound the Fisher information trace of every joint measurement on $t$ samples.<n>For the upper bound, we construct and analyze a nonadaptive tomography protocol based on a Gaussian joint measurement.
  arXiv  Detail & Related papers  (2026-09-09T17:48:41Z)
• Robust Average-Reward Markov Decision Processes: Minimax-Optimal Learning via Plug-in Reductions [51.50375419691955]
  Distributionally robust Markov decision processes provide a principled framework for sequential decision making under model uncertainty.<n>We study how many samples are necessary and sufficient to learn an $varepsilon$-optimal robust policy under the average-reward criterion.
  arXiv  Detail & Related papers  (2026-08-06T19:49:48Z)
• Sample Complexity of Multicalibration for Multilevel Properties [52.27687531970317]
  We study multicalibration for a sequence of $k$ properties in which each property is identifiable once the preceding properties are fixed.<n>Even with only polylogarithmically many binary groups, achieving multicalibration error $varepsilon$ requires $widetilde(varepsilon-(k+2))$ samples.
  arXiv  Detail & Related papers  (2026-08-04T23:38:54Z)
• Optimal Dimension-Free Sampling for Regularized Classification [56.72526267755301]
  We prove optimal sampling bounds achieving $(1pmvarepsilon)$-relative error for a broad class of Lipschitz continuous classification loss functions.<n>This includes important functions such as logistic and sigmoid loss, hinge loss, and ReLU loss, as prominent and popular representative examples.
  arXiv  Detail & Related papers  (2026-05-22T15:05:33Z)
• Order-Optimal Sequential 1-Bit Mean Estimation in General Tail Regimes [32.65125292684608]
  We propose a novel adaptive mean estimator based solely on randomized threshold queries.<n>Our estimator's sample complexity has an extra multiplicative $O(log(/))$ penalty.<n>For both threshold queries and more general interval queries, the sample complexity of any non-adaptive estimator must scale linearly.
  arXiv  Detail & Related papers  (2026-04-09T04:49:21Z)
• Optimal Unconstrained Self-Distillation in Ridge Regression: Strict Improvements, Precise Asymptotics, and One-Shot Tuning [61.07540493350384]
  Self-distillation (SD) is the process of retraining a student on a mixture of ground-truth and the teacher's own predictions.<n>We show that for any prediction risk, the optimally mixed student improves upon the ridge teacher for every regularization level.<n>We propose a consistent one-shot tuning method to estimate $star$ without grid search, sample splitting, or refitting.
  arXiv  Detail & Related papers  (2026-02-19T17:21:15Z)
• Optimal Lower Bounds for Online Multicalibration [9.852468478532652]
  We prove an $(T2/3)$ lower bound on expected multicalibration error using just three disjoint binary groups.<n>We then turn to lower bounds for the more difficult case of group functions that may depend on context but not on the learner's predictions.
  arXiv  Detail & Related papers  (2026-01-08T18:59:32Z)
• Sequential 1-bit Mean Estimation with Near-Optimal Sample Complexity [32.65125292684608]
  We study the problem of distributed mean estimation with 1-bit communication constraints.<n>Our estimator is $(epsilon, delta)$-PAC for all distributions with bounded mean ($-lambda le mathbbE(X) le lambda $) and variance ($mathrmVar(X) le sigma2$) for some known parameters.
  arXiv  Detail & Related papers  (2025-09-26T06:22:57Z)
• Settling the Sample Complexity of Model-Based Offline Reinforcement Learning [50.5790774201146]
  offline reinforcement learning (RL) learns using pre-collected data without further exploration. Prior algorithms or analyses either suffer from suboptimal sample complexities or incur high burn-in cost to reach sample optimality. We demonstrate that the model-based (or "plug-in") approach achieves minimax-optimal sample complexity without burn-in cost.
  arXiv  Detail & Related papers  (2022-04-11T17:26:19Z)
• The Sample Complexity of Robust Covariance Testing [56.98280399449707]
  We are given i.i.d. samples from a distribution of the form $Z = (1-epsilon) X + epsilon B$, where $X$ is a zero-mean and unknown covariance Gaussian $mathcalN(0, Sigma)$. In the absence of contamination, prior work gave a simple tester for this hypothesis testing task that uses $O(d)$ samples. We prove a sample complexity lower bound of $Omega(d2)$ for $epsilon$ an arbitrarily small constant and $gamma
  arXiv  Detail & Related papers  (2020-12-31T18:24:41Z)
• Optimal Robust Linear Regression in Nearly Linear Time [97.11565882347772]
  We study the problem of high-dimensional robust linear regression where a learner is given access to $n$ samples from the generative model $Y = langle X,w* rangle + epsilon$ We propose estimators for this problem under two settings: (i) $X$ is L4-L2 hypercontractive, $mathbbE [XXtop]$ has bounded condition number and $epsilon$ has bounded variance and (ii) $X$ is sub-Gaussian with identity second moment and $epsilon$ is
  arXiv  Detail & Related papers  (2020-07-16T06:44:44Z)
• Sample Complexity of Asynchronous Q-Learning: Sharper Analysis and Variance Reduction [63.41789556777387]
  Asynchronous Q-learning aims to learn the optimal action-value function (or Q-function) of a Markov decision process (MDP) We show that the number of samples needed to yield an entrywise $varepsilon$-accurate estimate of the Q-function is at most on the order of $frac1mu_min (1-gamma)5varepsilon2+ fract_mixmu_min (1-gamma)$ up to some logarithmic factor.
  arXiv  Detail & Related papers  (2020-06-04T17:51:00Z)
• Private Mean Estimation of Heavy-Tailed Distributions [10.176795938619417]
  We give new upper and lower bounds on the minimax sample complexity of differentially private mean estimation of distributions with bounded $k$-th moments. We show that $n = Thetaleft(frac1alpha2 + frac1alphafrackk-1varepsilonright)$ samples are necessary and sufficient to estimate the mean to $alpha$-accuracy under $varepsilon$-differential privacy, or any of its common relaxations.
  arXiv  Detail & Related papers  (2020-02-21T18:30:48Z)

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.