Overview

Generalization bounds estimate how well a model trained on finite data will perform on new examples, helping explain when learning is reliable. Common tools include VC and Rademacher complexity, covering numbers, PAC-Bayes and information-theoretic bounds, and algorithmic stability. For deep networks and modern settings, analyses increasingly use learned feature geometry, optimization dynamics, and structure in data, while extending guarantees to distribution shift, generative models, and dependent or distributed samples. A central tension is that broad worst-case guarantees can be too loose to be useful, whereas tighter, non-vacuous bounds often rely on assumptions about the algorithm, data geometry, or shift between training and test distributions.

Latest papers

172 papers12 papers in the last four weeks, down 50% on the four weeks before. 0.1% of all new papers.

Sep 30, 2026

Can Domain Generalization be Guaranteed in Small-Sample Learning?

The small-sample learning problem remains a fundamental challenge in machine learning because limited training data lead to unstable model estimation and generalization. Structural Risk Minimization (SRM) has long been regarded as a principled solution under the classical i.i.d. assumption. However, domain generalization (DG) violates this assumption, leaving the theoretical role of SRM in DG largely unexplored. To bridge this gap, we establish the first theoretical guarantees for SRM in DG under mild assumptions. Specifically, based on the concept of stability, we derive learning consistency and generalization error bounds and prove that these bounds become tight when the hypotheses satisfy the stability condition. Building upon this, under a specific hypothesis space assumption, we establish stability, learning, and generalization bounds for SRM. We further discuss the applicability of these bounds to deep learning. This work establishes theoretical foundations for SRM under distribution shifts and sheds light on the design of robust DG algorithms in small-sample scenarios.
Sep 30, 2026

How Many Samples Are Enough for Learning Across Domains?

Understanding the fundamental mechanisms of learning is essential for designing systems with strong generalization. Recent studies have shown that increasing the number of training domains, or enlarging the distribution shift among them, improves generalization when each domain contains sufficiently many data samples. However, the conditions under which the data samples can be considered sufficient remain unexplored. In this work, we fill this gap by establishing criteria for per-domain sample requirements based on the presented learning bounds. These criteria not only reveal an inverse linear scaling law between the number of training domains and the number of samples required per domain, but also explain the fundamental rationale behind the assumption of data sufficiency, thereby providing theoretical guidance for assessing the adequacy of existing datasets and constructing datasets. This differs from classical learning theory, as the number of samples required is highly dependent on the number of training domains. Additionally, we prove the close relationship between in-domain learning and out-of-domain generalization through the presented generalization bounds, and lastly discuss some key arguments.
Sep 29, 2026

Into the danger zone: stable extrapolation in high-dimensional function and operator learning

Out-of-distribution (OOD) generalization is a central challenge in scientific machine learning. We study regression problems in which the test distribution differs from the training distribution and ask: under what assumptions on the target function or operator is stable extrapolation possible, and how far beyond the training domain can one extrapolate? Existing theory controls the test error through additive penalties measuring the discrepancy between the training and test distributions. Such guarantees show robustness to small distribution shifts, but can be very pessimistic in comparison to OOD performance observed empirically. We identify classes of holomorphic functions and operators for which the OOD generalization error converges at algebraic rates even in the presence of large distribution shifts. This phenomenon stems from the increasing smoothness of higher-index coordinates, leading to what we term a `blessing of high dimensionality'. For learning with either polynomials, deep neural networks or deep neural operators, we derive explicit rates for arbitrary test measures supported on suitable domains and quantify how the admissible domain depends on the underlying regularity of the function or operator. Our extrapolation guarantees are independent of the test distribution, depending only on its support. We also present a series of numerical experiments across a range of functions and operators that support the main theoretical findings.
Sep 28, 2026

Preferent Compression Bounds Are Tight

The lack of rigorous safety and performance certificates remains a key bottleneck to the deployment of modern learning-based methods. Sample compression has recently emerged as a powerful tool for deriving such certificates, with particularly sharp bounds available for algorithms satisfying a so-called preference property -- also known as stability in the learning theory literature. These bounds find direct application across domains as different as the Scenario Approach, Pick-to-Learn, and Support Vector methods. However, whether they are tight has remained an open problem. In this paper we resolve this question affirmatively and show that the state-of-the-art bound for preferent compressions is provably tight. We establish this by exhibiting an explicit construction based on the uniform distribution and order statistics that attains the bound in the limit. Along the way, we also provide a considerably shorter and more accessible proof of this bound, requiring only elementary counting arguments and no infinite-dimensional duality.
Sep 28, 2026

Information-Theoretic Analysis of Next-Token Prediction under Markovian Data

We develop an information-theoretic framework for generalization in next-token prediction under temporally dependent data. We consider independent trajectories generated by finite-memory Markov processes and distinguish algorithmic dependence, quantified by mutual information, from temporal dependence, characterized by mixing. For cross-entropy loss, we derive an expected generalization bound using the Donsker--Varadhan variational representation and a McDiarmid-type concentration inequality for Markov chains. A refinement captures the joint effect of context length and temporal mixing through the mixing properties of the history-state process. We then extend the bound through a rate--distortion formulation, replacing mutual information with the minimum information rate required to represent the learned model within a prescribed distortion in the generalization gap, yielding informative guarantees for deterministic algorithms over continuous hypothesis spaces. For margin-based prediction, we derive explicit bounds for linear and self-attention next-token predictors via noisy low-dimensional compression, revealing the roles of context length, model complexity, sample size, margin, and temporal mixing. Experiments on TinyStories and ETTh2 show that longer contexts can reduce both training and test losses, but typically reduce training loss more, enlarging the generalization gap. A complementary ETTh2 analysis identifies an effective predictive-memory scale near 24 hours, with no statistically supported improvement beyond this scale, offering a plausible explanation for test-performance saturation at larger contexts.
Sep 25, 2026

Generalization behavior of OPTQ and the role of regularization

Large neural networks can be compressed by rounding or "quantizing" their weights to numbers that admit representations with fewer bits. One algorithm for quantization, OPTQ, progressively quantizes the weights of a neural network so that the squared quantization error on a specified calibration dataset is as small as possible. We study the performance of OPTQ and a variant algorithm, stochastic OPTQ, in a generalization setting and derive bounds for the expected squared error accrued by the algorithm when a test point is drawn from a fixed distribution. We prove two results. One result relates the generalization error to the error on a calibration dataset comprising independent samples from the same distribution as the test distribution. The other result bounds the generalization error of stochastic OPTQ for all sufficiently nice distributions, regardless of the calibration dataset. In both of these results, the regularization term λλ plays an important role. We use insights from these results to make a new recommendation for the choice of λλ and see that this choice of λλ preforms favorably in experiments when compared to prior recommendations in the literature.
Sep 23, 2026

Even Sharper Bounds for Transductive Learning and Its Applications

We introduce Sharper Transductive Local Complexity (STLC), a localized complexity method for transductive learning under uniform sampling without replacement. The construction starts from a Bernstein-type concentration inequality for the supremum of the test--train empirical process. Its proof uses the modified log-Sobolev inequality for the swap walk and a two-parameter entropy closure. A peeling argument with a surrogate localization functional then gives excess-risk bounds with the same fixed-point and confidence terms as the classical inductive local Rademacher-complexity bounds, without the additional logarithmic confidence factor in earlier transductive results. For realizable learning over a binary class of VC dimension \dVC\dVC, with training size mm, test size uu, and u≥m≥\dVCu\ge m\ge\dVC, STLC yields \cO{\dVClog⁡(me/\dVC)/m}\cO\{\dVC\log(me/\dVC)/m\}. This matches the standard inductive rate and, when m≥9m\ge9, is within a logarithmic factor of the transductive minimax lower bound of order \dVC/m\dVC/m. For transductive kernel learning, STLC gives a spectrum-adaptive excess-risk bound without the multiplicative imbalance factors appearing in the earlier local-complexity bound.
Sep 15, 2026

Sharp margin-based generalization bounds for realizable SVM

Let the exact homogeneous hard-margin support vector machine be trained on mm independent observations from a Borel probability law on a real Hilbert space. We prove that, with score zero counted as an error, there is a universal numerical constant CC such that \Pp(γm>0,\Risk(um)>Cm(Km+log⁡1δ))≤δ.\Pp\left( γ_m>0,\quad \Risk(u_m)> \frac{C}{m} \left( K_m+\log\frac1δ \right) \right) \le δ. Here γmγ_m is the empirical homogeneous margin, umu_m is the exact minimum-norm unit-margin separator, rmr_m is the largest training radius, and Km:=rm2\normum2=rm2/γm2K_m:=r_m^2\norm{u_m}^2=r_m^2/γ_m^2 on {γm>0}\{γ_m>0\}. The proof is driven by a deterministic deletion problem. Given vectors x1,…,xnx_1,\ldots,x_n in the unit ball, delete a set BB of constraints and let uBu_B be the closest point to the origin that satisfies every retained unit-margin constraint. Suppose that \normuB2≤k\norm{u_B}^2\le k and that every deleted vector has nonpositive score under uBu_B. We prove that a family of such deletion sets of cardinality qq has size at most exp⁡(8k+2q)\exp(8k+2q). The conceptual step is an exact identity obtained from the KKT representation of uBu_B. For a random deletion set, the identity converts the mean squared spread of the separators into a weighted sum of score deficits. It therefore forces a coordinate whose deletion status separates the two conditional means by a quantitatively large amount. Revealing that coordinate decreases the conditional separator variance enough to control the binary entropy of the split. An entropy induction gives the deletion count, and an exact factorial ghost-sample identity converts that count into the stated high-probability SVM bound.
Sep 10, 2026

General Quantification of Covariate and Concept Shifts

Generalization under distribution shift remains a core challenge in modern machine learning, yet existing learning bound theory is limited to narrow, idealized settings and is non-estimable from samples. In this paper, we bridge the gap between theory and practical applications. We first show that existing definition of concept shift breaks when the source and target supports mismatch. Leveraging entropic optimal transport, we propose a key notion: γ∗ ⁣γ^{*}\!-concept shifts, and derive a general error bound unifying covariate and γ∗ ⁣γ^{*}\!-concept shifts, which applies to broad loss functions, label spaces, and stochastic labeling. We further develop estimators for these shifts with concentration guarantees, and the DataShifts algorithm, which can quantify distribution shifts and estimate the error bound in most applications - a rigorous and general tool for analyzing learning error under distribution shift.
Sep 10, 2026

Symmetry-quotient Flatness and Generalization

This paper develops a theorem-level pipeline in symmetry-quotient settings: quotient linear stability implies quotient flatness, quotient flatness implies input smoothness, and input smoothness yields generalization under local covering assumptions. Flatness is often associated with generalization, and Stochastic Gradient Descent (SGD) is frequently viewed as implicitly biased toward flat solutions. However, standard flatness measures are typically defined in the raw parameter space and are therefore not invariant under function-preserving symmetries such as positive rescaling. We develop a symmetry-aware theory of quotient flatness, quotient linear stability, input smoothness, and generalization on quotient spaces of neural-network parameters. For square loss and models equipped with function-preserving group actions, we define quotient flatness as the trace of the Hessian of the empirical loss on the regular quotient manifold. We show that quotient flatness controls input smoothness through a quotient-space analogue of the flatness-to-smoothness argument. We also prove that one-step mean-square quotient linear stability of the linearized SGD dynamics implies an explicit quotient-flatness bound in terms of the batch size and learning rate, and extend this analysis to higher-order tensor moments. Finally, under local covering and boundedness assumptions, we derive population generalization bounds in terms of quotient flatness and, consequently, in terms of quotient linear stability.
Sep 8, 2026

Length Generalization for Transformers via Compression

Recent advancements in transformer length generalization theory enable us to reliably predict when a transformer can learn to solve a task. In particular, the C-RASP hypothesis (a formalized version of the so-called RASP-l conjecture) posits that transformers length-generalize on a task if and only if a solution is expressible in the C-RASP language. While this hypothesis has strong empirical validation, theoretical problems arise from the fact that no computable length generalization bounds exist for C-RASP, alongside the discovery of seemingly contradictory experiments. To address these problems, we refine the C-RASP hypothesis utilizing the recently-proposed fragments C-RASP+ and C-RASP1. These fragments have computable length generalization bounds, though in the worst case requiring an extremely large (double exponential) sample size. It is an open question whether these sample size bounds are tight. In this paper, we resolve this open question by providing an exponentially tighter bound. In doing so, we show a polynomial length generalization bound for transformers if we adopt compressed strings, via a novel connection to power words. As an application, we show how this yields a fine-grained analysis of the C-RASP conjecture that resolves contradicting experimental evidence against it.
Sep 7, 2026

A Theoretical Analysis of Generalization Dynamics in Neural Networks under Gradient Descent with Weight Decay

Understanding generalization remains a central challenge in machine learning because it requires jointly considering data, architecture, and training dynamics. In this paper, we develop a theoretical framework that characterizes how these factors jointly shape generalization performance throughout training. More precisely, we study a broad class of neural networks trained under the ℓ2\ell^2 loss by gradient descent (GD) with weight decay, and prove the convergence of GD to a neighbourhood of the global minimizers of the empirical loss. By partitioning the space based on the input data, we then decompose the population error into data error, optimization error, and prediction variation error, and bound them separately. In particular, for the prediction variation error, which measures the oscillations of the learned function, we propose (local) approximate homogeneity and derive explicit cellwise and layerwise bounds for its evolution along the training trajectory. These bounds yield two important implications: a necessary condition of improved generalization explains differences in layerwise generalization behavior; a sufficient condition describes delayed generalization and provides a theoretical characterization of grokking.
Sep 1, 2026

On the Reliability of Generative Augmentation: A Wasserstein-Based Theoretical and Empirical Study

Generative data augmentation is widely used to mitigate class imbalance, yet its theoretical effect on downstream generalization remains poorly understood. In this work, we develop a statistical framework for conditional generative augmentation and analyze its impact on classification risk. We formalize augmentation as a distribution-mixing process and show that the resulting risk distortion is controlled by both the augmentation strength and the class-conditional Wasserstein discrepancy between real and generated distributions. We further derive a capacity-dependent generalization bound based on Rademacher complexity, revealing an explicit trade-off between hypothesis complexity, augmentation intensity, and generative fidelity. Empirically, we evaluate the framework on binary and multiclass imbalanced classification tasks using Conditional GAN and Conditional WGAN-GP augmentation. Across datasets, CWGAN-GP consistently achieves lower Wasserstein discrepancies than CGAN, indicating improved distributional fidelity. However, improved fidelity does not necessarily translate into superior classification performance, with classical oversampling methods often remaining competitive. These findings support the central theoretical prediction that augmentation reliability is governed by distributional approximation error rather than predictive performance alone. Overall, this work establishes generative augmentation as a distributional perturbation process whose reliability can be quantified through Wasserstein-based measures and supported by finite-sample generalization guarantees. The proposed framework provides a principled foundation for evaluating synthetic data quality beyond classification accuracy alone.
Sep 1, 2026

Exact Risk-Complexity Laws for Projective Boundaries in Scenario Optimization and Distribution-Free Certification

Scenario optimization, conformal prediction, and related distribution-free certification methods use finite samples to construct decisions or prediction sets with violation-risk guarantees for fresh observations. In several classical settings, the conditional violation risk follows an exact beta law, whose tail has a beta-binomial representation and whose parameter is a support, calibration, or compression dimension. This paper identifies the deterministic boundary mechanism behind these formulas and derives the corresponding law when the observed boundary size is random. A decision rule is represented by an acceptance set for future observations, together with a boundary map selecting the sample points responsible for that set. The resulting pair is called a {\em proper projective boundary scheme} when held-out samples are accepted precisely if the full-sample boundary is retained, and accepted non-boundary samples can be deleted without changing that boundary. For every such scheme, the conditional law of the violation risk given the observed boundary size is determined by the boundary's cross-sample complexity profile. A stable profile yields the usual beta law, whereas a varying profile produces an exact profile correction. The framework covers scalar order-statistic calibration, support-reconstructive scenario programs, cascaded support-removal certificates, coordinatewise envelopes, and Pareto-frontier calibration with vector scores. It also yields conditional probabilistic certificates and a no-go result explaining why observed complexity alone is insufficient.
Aug 31, 2026

Generalization as a robust performance property of learning-enabled dynamical systems

By focusing on algorithmic stability as a means of establishing out-of-sample bounds, we provide a system-theoretic interpretation of generalization in learning-enabled dynamical systems arising in data-driven optimization and feedback control approximation. Given two neighboring datasets, we specifically model sample replacement as an exogenous disturbance acting on a sensitivity system, while the incremental behavior of the data-dependent operator is encoded through an integral quadratic constraint. By relying on dissipativity arguments, we establish a matrix inequality-based certificate and a uniform stability bound that separates the one-sample sensitivity of the learned operator, and an algorithm-dependent dynamical gain. The latter can then be optimized, offering a tractable tool for certifying and comparing generalization capabilities of learning dynamics. We show that our results recover classical ones for gradient descent, apply naturally to momentum-based methods such as heavy-ball and Nesterov acceleration, and extend to data-driven control.
Aug 25, 2026

The Sharp Tail of Uniform Stability

Uniform stability controls how much one training example can change the loss at any test point. A new logarithmic-free upper bound shows that a γγ-uniformly stable algorithm with loss in [0,L][0,L] has generalization gap at most O(γlog⁡(1/δ)+Llog⁡(1/δ)n)O \left(γ\log(1/δ) +L\sqrt{\frac{\log(1/δ)}{n}}\right) with probability 1−δ1-δ. Whether an actual bounded-loss learning algorithm can realize the linear dependence on log⁡(1/δ)\log(1/δ) has remained open. The known construction realizes it only for auxiliary weakly dependent random variables whose pointwise range grows with nn. The known learning lower bound holds only at constant probability. We close this gap. For every nn, stability level γγ, and loss bound LL, we construct one deterministic γγ-uniformly stable learning problem whose tail satisfies, simultaneously for 1≤p≤cn1\le p\le c n, P(R(AS)−RS(AS)≥c′min⁡{L,γp+Lp/n})≥e−p.\mathbb P \left( R(A_S)-R_S(A_S) \ge c'\min \left\{L,γp+L\sqrt{p/n}\right\} \right)\ge e^{-p}. The construction is ordinary bounded absolute-loss regression with constant labels. Its key is a multiscale collection of rare Rademacher features. A coordinatewise ramp is stable in sup norm, while an odd symmetrized maximum converts a unique extreme feature into a gap of order γpγp without violating the loss bound. Geometrically spaced ramps put all confidence levels into the same problem. Together with the logarithmic-free upper bound, this determines the optimal high-probability and moment dependence of uniform stability up to universal constants.
Aug 18, 2026

Tight Bounds for Data-driven Multiple Hyper-parameter Tuning with Structured Loss Function

Data-driven algorithm design frames hyperparameter tuning as a statistical learning problem, but establishing generalization guarantees remains challenging due to the implicit, non-smooth dependence of model performance on hyperparameters. Existing multi-dimensional bounds under piecewise-polynomial assumptions remain theoretically loose and lack comprehensive lower bounds. We resolve this by establishing tight pseudo-dimension bounds for multi-dimensional data-driven tuning. First, we refine the learning-theoretic upper bound using real algebraic geometry; by analyzing invariant connected sign cells during block elimination rather than isolated sign vectors, we avoid topological over-counting to derive strictly sharper sample complexities. Second, we present a multi-regime lower-bound framework that disentangles combinatorial and algebraic capacities. By constructing shattered problem instances across distinct regimes, we prove our upper bounds are tightly saturated. Finally, we extend our topological framework to accommodate general bi-level validation-loss tuning and broader semi-algebraic applications.
Aug 17, 2026

Operator-Theoretic Generalization Bounds for Multitask Deep Learning

We develop operator-theoretic generalization bounds for deep multi-output function classes by representing network layers as Koopman composition operators on vector-valued reproducing kernel Hilbert spaces. In vector-valued Sobolev RKHSs, we derive Rademacher complexity bounds for invertible and width-expanding injective architectures. The estimates separate the output-coupling contribution, represented by the trace of the task matrix, from the layerwise operator norms, Sobolev symbol ratios, determinant factors, and restriction constants generated by the linear maps. We then analyze a distinct one-dimensional Brownian/Cameron--Martin regime. Using the exact anchored derivative-norm characterization of the vector-valued Brownian RKHS, we obtain layerwise bounds for domain-preserving scalar linear maps and anchored diffeomorphic activations; the corresponding factors scale as ∣Wl∣1/2|W_l|^{1/2} and ∥σl′∥∞1/2\|σ_l'\|_\infty^{1/2}, respectively, and do not involve Sobolev smoothness exponents. Because the Sobolev and Brownian results concern different hypothesis spaces, neither is asserted to dominate the other uniformly. We additionally formulate shared operator learning across tasks, prove a finite-rank representer theorem, derive the exact finite-dimensional problem for squared loss, and establish a target-transfer bound when the learned operator is obtained independently of the target sample. Synthetic and MNIST studies examine stabilized Sobolev-inspired and Brownian-inspired complexity proxies; these empirical proxies are not evaluations of the proved bounds for rank-deficient architectures.
Aug 12, 2026

Drift and Dependence: Layer-wise Information-Theoretic Bounds for Replay-Based Continual Learning

Continual learning must absorb new tasks without erasing old ones, and replay---mixing a small buffer of past examples into current training---is among the most effective remedies for catastrophic forgetting. Yet its generalization behavior is shaped by two coupled effects that existing analyses fold into a single hypothesis-level quantity: finite memory replaces each past distribution with an empirical proxy, and repeated reuse couples the buffer, the current data, and the final hypothesis through a shared optimization trajectory. We develop a layer-wise information-theoretic framework that separates these effects at every depth. Our main result decomposes the expected generalization gap into a replay-induced representation drift and an optimization-dependence term, the latter further resolved into stability, plasticity, interaction, and residual-coupling components. Two refinements make the framework operational. A Wasserstein relaxation of the drift term, valid under support mismatch, yields a depth-dependent drift--sensitivity trade-off whose minimizer identifies which interior layer to stabilize. An SGLD instantiation of the optimization term reduces it to a trajectory-level log-determinant budget, exposing a curvature-aware gradient-alignment statistic that serves as an online diagnostic of task-wise forgetting. Controlled and benchmark experiments confirm the predicted memory scaling, the interior funnel, and the alignment signal's link to forgetting.
Aug 11, 2026

Optimistic Rates for Multiclass PAC Learning

Worst-case multiclass bounds do not become smaller when the best classifier is already nearly correct: what is missing is an optimistic rate, a guarantee whose fluctuation scales with the oracle risk itself. For a class of Natarajan dimension dNd_N and Daniely-Shalev-Shwartz dimension dDSd_{DS}, the optimal excess risk is known at the two endpoints (dDS/nd_{DS}/n realizable, dN/n+dDS/n\sqrt{d_N/n}+d_{DS}/n agnostic [HMZ24, CEH+26, Pab26]) and open in between. We close the gap: at every fixed oracle risk L⋆L^\star, the optimal excess risk is Θ~(L⋆dN/n+dDS/n)\widetildeΘ(\sqrt{L^\star d_N/n}+d_{DS}/n), uniformly in the alphabet size, attained by a learner that knows neither L⋆L^\star nor the confidence level. The upper bound composes the cover-menu-compression architecture of [CEH+26], at the realizable rate of [Pab26], with a new comparator-facing relative compression theorem: a size-kk compression rule that empirically dominates a comparator hh has population risk at most L(h)+O(L(h)Γ+Γ)L(h)+O(\sqrt{L(h)Γ}+Γ) with Γ=(klog⁡n+log⁡(1/δ))/nΓ=(k\log n+\log(1/δ))/n, without stability; this transfers the comparison principle of the sharp binary theory [MQZ26] while discarding its Boolean-cube geometry, which does not lift to multiclass labels. The lower bound forces both terms using one class and one distribution at every fixed L⋆L^\star, by a pair-Assouad scheme calibrated to L⋆L^\star and a fiber argument on the pseudo-cubes underlying the Natarajan-versus-DS separation of [BCD+22]. Both theorems extend to list learning: against the best rr-tuple of hypotheses, the same architecture and the same two engines yield an optimistic rate and a lower bound of the same shape, forcing the fluctuation term that [Pab26] expected to be necessary against list comparators, and removing the factor rr from the known realizable list lower bound.
Aug 10, 2026

Logarithmic-Free Moment and Generalization Bounds for Uniformly Stable Algorithms

Uniform stability is a classical tool for controlling the generalization error of a learning algorithm. Bousquet, Klochkov, and Zhivotovskiy (2020) showed that the problem can be reduced to a moment inequality for a sum of weakly interacting functions of independent random variables. Their bound contains an additional factor log⁡n\log n, and they asked whether this factor can be removed. We answer this upper-bound question affirmatively. More specifically, let Z=(Z1,…,Zn)Z=(Z_1,\ldots,Z_n) have independent coordinates and let gi(Z)g_i(Z) satisfy E[gi(Z)∣Z−i]=0, ∣E[gi(Z)∣Zi]∣≤M, for every i=1,…,n,\mathbb E[g_i(Z)\mid Z_{-i}]=0, \ \left| \mathbb E[g_i(Z)\mid Z_i]\right|\le M, \ \text{for every } i = 1, \dots, n, where Z−iZ_{-i} denotes all coordinates except ZiZ_i. Assume additionally that changing any coordinate ZjZ_j, j≠ij\neq i, changes gig_i by at most ββ, we prove that, for every p≥2p\ge2, for every p≥2p\ge2, ∥∑i=1ngi(Z)∥p≤16pnβ+M2pn.\left\| \sum_{i=1}^n g_i(Z)\right\|_p \le 16pnβ+M\sqrt{2pn}. This removes the log⁡n\log n factor from the previous bound and matches the lower bound of Bousquet, Klochkov, and Zhivotovskiy up to universal constants in the range covered by their construction. Our proof first establishes the required estimate on the Rademacher cube, then transfers it to arbitrary product distributions by a two-copy randomization argument.
Aug 6, 2026

An Optimal Agnostic PAC Algorithm

Let H⊆{−1,+1}XH\subseteq\{-1,+1\}^X be a class of finite VC dimension d≥1d\ge1. Writing LL for the binary risk and L∗=min⁡h∈HL(h)L^*=\min_{h\in H}L(h), we construct a learner achieving the statistically optimal risk bound: from an i.i.d.\ sample of size nn, for every 0<δ≤1/20<δ\le 1/2, with probability at least 1−δ1-δ, L(h^)≤L∗+7⋅108(L∗(d+log⁡(1/δ))n+d+log⁡(1/δ)n).L(\widehat h) \le L^*+ 7\cdot10^8\left( \sqrt{\frac{L^*(d+\log(1/δ))}{n}} +\frac{d+\log(1/δ)}{n} \right). This settles the sample complexity of agnostic PAC learning up to universal constants at every fixed L∗L^*, matching the lower bounds of Devroye, Györfi, and Lugosi [A Probabilistic Theory of Pattern Recognition, Springer, 1996].
Aug 5, 2026

Variational Bounds for Perceptron Learning from Structured Data

We introduce a variational approach to a finite-temperature continuous-spin perceptron trained on a Gaussian mixture. The model allows for a broad class of concave utilities and log-concave separable prior measures on the spins. By combining the interpolation method with log-concavity and concentration estimates, we derive lower and upper minimax variational bounds for the limiting quenched pressure. Remarkably, the two bounds differ only in the order of optimization of two variational parameters, while all remaining extrema are controlled by the concave--convex structure of the variational potential. Whenever the two optimizations commute, the two bounds match and identify the solution of the model. The same potential yields the fixed-point equations as stationarity conditions and provides a unified route to the computation of the ground-state energy, training loss, and generalization error.
Jul 30, 2026

Generalization Bounds on Optimal Control for Transformer Training and Wasserstein Distributional Robustness

We derive finite-sample generalization bounds for Transformers trained with dynamic programming recursions. Building on the doubly lifted, measure-valued formulation of Transformer dynamics, we view data sets as probability laws on pairs of empirical input-output measures, allowing us to interpret the training problem as a finite-horizon Markovian control problem. We then analyze a quantized model, derived by quantizing the state, action, and measure-state spaces, and derive explicit finite-sample generalization bounds using concentration inequalities for empirical laws on finite metric spaces together with a Lipschitz stability estimate for the value function. These bounds are transferred to the base model at the cost of an explicit approximation error. Finally, we show that the same machinery yields a distributionally robust control formulation of the training problem, connecting Transformer generalization to Wasserstein distributionally robust optimization.
Jul 29, 2026

Tight Generalization Bound for AdaBoost

In this paper we show that the generalization error of AdaBoost is Θ(dln⁡(nγ2/d)nγ2+ln⁡(1/δ)n)Θ\big(\tfrac{d\ln(nγ^{2}/d)}{nγ^2}+\tfrac{\ln(1/δ)}{n}\big), where γγ is the advantage guaranteed by the weak learner, dd is the VC-dimension of the class containing the weak hypotheses, nn is the sample size, and δδ is the confidence parameter. The contribution of this paper is the upper bound; the matching lower bound follows from prior work. The upper bound proof follows by combining the known fact that AdaBoost outputs a voting classifier whose voting function has zero empirical γ/2γ/2-margin loss with what is, to the best of our knowledge, a new margin-based generalization bound for voting classifiers.
Jul 21, 2026

H2(Ω)H^2(Ω) A Priori Error Bounds for Boundary-Adapted PINNs Solving Elliptic Dirichlet Problems

We study the numerical resolution of uniformly elliptic Dirichlet boundary value problems (BVPs) by minimization of the PDE residual over a hypothesis space of Neural Networks made to satisfy exactly the boundary conditions through multiplication by a user-defined distance-to-boundary approximation ρρ. While an L2(∂Ω)L^2(\partialΩ) boundary penalty cannot generally guarantee H2(Ω)H^2(Ω) error control, we show that exact enforcement can, provided ρρ is chosen appropriately. More precisely, we identify a sharp sufficient condition on ρρ preserving the regularity needed for H2(Ω)H^2(Ω) error control, provide practical constructions of such ρρ for general smooth domains ΩΩ, and refer to the resulting networks as boundary-adapted. For boundary-adapted Rectified Quadratic Unit (ReQU) and hyperbolic tangent (tanh) networks, we then derive explicit high-probability H2(Ω)H^2(Ω) convergence rates by combining approximation-theoretic and statistical-learning arguments, making explicit the dependence on ρρ. Numerical experiments support our theoretical results, showing that appropriate choices of ρρ improve accuracy and convergence, while poorly chosen distance functions can substantially degrade the solution. Our proof also yields new VC-dimension bounds for hypothesis spaces of higher-order derivatives of ReQU and tanh networks, together with new approximation bounds for shallow ReQU networks in higher-order Sobolev norms, all of which are of important independent interest.
Jul 17, 2026

Hierarchical Domain Generalization

We study hierarchical domain generalization as a problem of extrapolation from finite observed regions to an entire instance space, replacing i.i.d. sampling with arbitrary domain hierarchies. We show that the central obstruction is not only the complexity of the hypothesis class, but the train/test domain partition through which evidence is revealed. In particular, no matter how small the class or how large the training size, some partition makes generalization fail for some target. These results suggest that modern generalization theory must treat domain structure as a first-class object.
Jul 9, 2026

Generalization Theory for Through-the-Wall Radar Human Activity Recognition

Through-the-wall radar (TWR) human activity recognition (HAR) is important for non-line-of-sight indoor sensing, security monitoring, and emergency rescue. However, structured distribution shifts caused by person variation, observation-view variation, and wall-condition variation severely degrade recognition generalization, while the origin of the target-domain error still lacks a rigorous theoretical explanation. To address this issue, a generalization-analysis framework for TWR HAR is proposed in this paper. First, models for indoor human kinematics, TWR echo generation, radar image formation, feature representation, and bounded-weight neural networks are established within a unified source-to-target learning formulation. Then, the source risk, target risk, empirical risk, and admissible physical domain descriptor are defined, and a unified target-domain generalization bound is derived. Next, the structured shift term is decomposed into cross-person, cross-view, and cross-wall components, and the bound-tightening effects of physical low-dimensional representations, multi-source training, and parameter-space coverage are analyzed. Simulated and measured experiments jointly support the resulting theoretical analysis and illustrate its application value.
Jul 9, 2026

An exact information theory of generalization phase transitions in Bayesian diffusion models

How diffusion models circumvent the curse of dimensionality to learn complex distributions over high dimensional spaces from a finite training set, instead of memorizing it, remains a fundamental mystery. To address this, we introduce analytically tractable Bayesian information restricted diffusion (BIRD) models, in which each pixel observes restricted information about noisy data. A BIRD model time-reverses diffusion by inferring which past training sample produced its current restricted observation using the Bayesian posterior. This model class generalizes existing analytical diffusion models that use spatially local information restriction. We show that spatially local BIRD models closely approximate trained diffusion models early in training, across different architectures such as UNets and DiTs. Under minimal assumptions on the data distribution, we identify an information-theoretic phase boundary between memorization and generalization in the joint space of amount of training data, time in the reverse generative process, and amount of information restriction: a BIRD model memorizes when the mutual information between its restricted noisy observations and the training data exceeds the log number of training points, and it generalizes otherwise. Experiments across a range of datasets confirm our theoretically predicted location for the transition. We find that generation proceeds near the edge of memorization: both spatially local BIRD models and early-training diffusion models track the memorization-generalization phase boundary by increasingly restricting information over time. Overall, our results reveal a fundamental role for information restriction in generative AI to circumvent the curse of dimensionality.
Jul 8, 2026

Any-Dimensional Learning by Sampling

Many machine learning models are defined for inputs of different sizes, such as point clouds containing different numbers of points, sequences of tokens of different lengths, and graphs on different numbers of nodes. Such models are trained on finitely many examples of necessarily limited sizes. How well do these models generalize from inputs of small size to larger inputs of size not seen during training? Furthermore, evaluating such models on large inputs is often expensive. How can we sketch large inputs to obtain smaller ones on which the model takes similar values? At the heart of both questions is the need to compare inputs of different sizes and to approximate large inputs by small ones. We present a unified approach to address these questions by using random sampling maps to compare inputs of different sizes. The sampling maps we consider are generalizations of sampling with replacement, random binning, and species sampling. We characterize the application domains in which each type of sampling is appropriate in terms of the symmetries and relations between problem instances of different sizes in the domain. Our framework yields explicit generalization and sketching rates for function classes continuous with respect to a chosen notion of sampling, encompassing large families of functions defined on sequences, graphs, and tensors of different sizes. Specific examples include moment polynomials on measures, homomorphism densities and numbers of graphs, permutation-invariant transformers, and graph neural networks.