New bounds for LDP with heterogeneous privacy levels guaranteeing high probability of accuracy.
problem Statistical estimation under LDP with users having varying privacy levels.
method Developed finite sample upper bounds in ℓ_2-norm with high probability, complemented by lower bounds.
result Optimal guarantees for heterogeneous LDP in terms of probability and constants.
Sharp bounds for high-probability estimation of discrete distributions.
problem Estimating discrete distributions with high probability under χ2-divergence. method Sharp upper and lower bounds for the classical Laplace estimator, and characterization of minimax high-probability risk for any estimator.
result Sharp bounds for high-probability estimation of discrete distributions can be achieved through a simple smoothing strategy.
New approach for online learning with adaptive adversaries, simpler and more effective.
problem Online learning with adaptive adversaries, especially in bandits and MDPs.
method Uses standard unbiased estimators and a simple increasing learning rate schedule, aided by logarithmically homogeneous self-concordant barriers and strengthened Freedman's inequality.
result First high-probability regret bounds for adversarial bandits and MDPs, resolving open problems.
New algorithm reduces high-probability regret for time-varying feedback graphs.
problem High-probability regret bounds for adversarial bandits with time-varying feedback graphs.
method Online mirror descent framework with innovative techniques for pessimistic loss estimators.
result Achieves optimal high-probability regret bound for general and weakly observable graphs.
New method bounds stochastic subgradient methods with heavy-tailed noise.
problem Bounding stochastic subgradient methods under heavy-tailed noise.
method Clipped version of projected stochastic subgradient method.
result Near optimal any-time and finite horizon bounds for averaging schemes.
The paper provides bounds for LSA with fixed stepsizes under random estimates.
problem Analyzing the performance of LSA algorithms with fixed stepsize.
method Non-asymptotic analysis based on new results about matrix moments and high probability bounds.
result Derives high probability bounds on LSA performance under weaker conditions than previous works.
Unified high-probability regret bounds for online convex optimisation with randomised gradient estimators.
problem Online convex optimisation with randomised gradient estimators for ℓq-Lipschitz losses. method FTRL with randomised two-point finite-difference gradient estimators based on cone-measure sampling from ℓr-spheres. result Unified high-probability regret bounds for all p,q,r∈[1,∞]. The study optimizes distribution estimation from samples with relative entropy error, adapting to sparse distributions.
problem Estimating discrete distributions with high-probability accuracy in relative entropy.
method Analysis of Laplace estimator and confidence-dependent smoothing techniques, including data-dependent smoothing.
result Optimal high-probability risk bounds for various estimators, including a new data-dependent smoothing method.
Improved bounds for discrete probability distribution estimation under the ℓ∞ norm.
problem Estimating discrete probability distributions under the ℓ∞ norm with improved bounds.
method Minimax bounds in expectation and high-probability tail bounds.
result Resolved open questions posed in Kontorovich and Painsky (JMLR, 2025), including a fully empirical tightest risk bound and identifying the worst-case extremal distribution.
Method solves complex optimization problems with high probability bounds.
problem Nonlinear equality constrained stochastic optimization problems.
method Step-search sequential quadratic programming method.
result High-probability bound on iteration complexity for first-order stationarity.
We derive concentration inequalities for differentially private median and mean estimators building on the "Propose, Test, Release" (PTR) mechanism introduced by Dwork and Lei (2009). We introduce a new general version of the PTR mechanism that allows us to derive high probability error bounds for differentially privat…
High-probability bound for distributed stochastic approximation tracking error.
problem Analyzing the convergence of distributed stochastic approximation schemes.
method Analysis using ODE approach to stochastic approximation.
result High probability bound for tracking error between iterates and limiting differential equation.
Wavelet-based online learning adapts to noisy Besov spaces with high probability.
problem Minimizing integrated squared error in Besov spaces with noisy observations.
method Adaptive wavelet-based online learning algorithm that dynamically adjusts to gradient noise.
result Achieves minimax-optimal integrated squared error with high probability.
New bounds show faster convergence for learning algorithms.
problem Improving risk bounds for learning algorithms.
method Using algorithmic stability and common assumptions like Polyak-Lojasiewicz condition, smoothness, and Lipschitz continuity.
result Achieves convergence rate of O(log2(n)/n2) with high probability. RS-NSGD improves SGD convergence for heavy-tailed noise.
problem Nonconvex optimization with heavy-tailed noise.
method Integrates direction normalization into subspace updates.
result Achieves better oracle complexity than full-dimensional normalized SGD.
Dynamic Vocabulary Pruning stabilizes LLM training by removing low-probability tokens.
problem Training Large Language Models (LLMs) with Reinforcement Learning (RL) causes numerical divergence between inference and training.
method Dynamic Vocabulary Pruning (DVP) constrains the RL objective to a safe vocabulary that excludes low-probability tokens.
result DVP stabilizes training by reducing systematic bias introduced by the extreme tail of the token distribution.
New bounds improve generalization in machine learning with high probability.
problem Erratic behavior of KL divergence limits practical applications.
method Replaced KL divergence with Wasserstein distance for better bounds.
result Proved high probability generalization bounds for i.i.d. and non-i.i.d. data.
The paper analyzes the robustness of a minimum ℓ2 interpolator in high-dimensional linear regression.
problem Analyzing the robustness of a minimum ℓ2 interpolator in high-dimensional linear regression. method The paper analyzes the interpolator with minimal ℓ2-norm in a general high-dimensional linear regression framework, proving bounds on prediction loss. result The paper shows that the prediction loss of the interpolator is bounded by (∥β∗∥22rcn(Σ)∨∥ξ∥2)/n with high probability, revealing a transition in rates. The study tightens bounds on binomial probabilities and minimums using KL-divergence.
problem Tightening bounds on binomial probabilities and minimums of i.i.d. Binomials.
method Applied Sanov's theorem to derive upper and lower bounds on binomial tail probabilities and minimums, expressed in terms of KL-divergence.
result High probability upper and lower bounds on the minimum of i.i.d. Binomial random variables, finite sample, asymptotically tight.
This work addresses the problem of regret minimization in non-stochastic multi-armed bandit problems, focusing on performance guarantees that hold with high probability. Such results are rather scarce in the literature since proving them requires a large deal of technical effort and significant modifications to the sta…
This paper addresses missing covariates in stochastic linear bandits, providing a high-probability regret bound.
problem Effect of missing covariates on regret in stochastic linear bandit algorithms.
method Proposes an algorithm that provides a high-probability upper bound on regret in terms of covariate sampling probabilities.
result Regret degrades due to missingness by at most ζmin2, where ζmin is the minimum probability of observing covariates. New algorithms achieve high-probability parameter-free regret in online convex optimization with heavy-tailed data.
problem Achieving high-probability parameter-free regret in online convex optimization with heavy-tailed data.
method Developed new regularization techniques to handle exponentially large iterates and heavy-tailed subgradients.
result Achieved regret bound of O(∥u∥T1/plog(1/δ)) with high probability for subgradients with bounded pth moments. The paper analyzes Adam and SGD in nonstationary optimization, revealing tradeoffs between noise and drift.
problem Analyzing Adam and SGD in nonstationary optimization problems.
method Theoretical analysis of Adam and SGD under non-stationary stochastic objectives, separating two regimes.
result Characterizes the tradeoff between noise and drift in Adam and SGD, revealing when adaptive step-sizing is beneficial or harmful.
The Gibbs algorithm's generalization error is bounded, improving with prior volume in low temperatures.
problem Bounding the generalization error of the Gibbs algorithm in low temperature regimes.
method Analyzes the Gibbs algorithm's performance, extending known high-temperature bounds to low-temperature scenarios.
result With high probability, the generalization error decreases with the total prior volume of similar hypotheses.
Algorithmic stability is a classical approach to understanding and analysis of the generalization error of learning algorithms. A notable weakness of most stability-based generalization bounds is that they hold only in expectation. Generalization with high probability has been established in a landmark paper of Bousque…
New bounds using samplewise evaluated CMI for deep neural networks.
problem Improving generalization bounds for deep neural networks.
method Introduced a new family of information-theoretic generalization bounds using samplewise evaluated conditional mutual information (CMI).
result The new bounds can be tighter than previous ones for deep neural networks.
Paper develops a TR-SSQP method for noisy optimization with heavy-tailed noise.
problem Optimization problems with stochastic objectives and heavy-tailed noise.
method Trust-Region Stochastic Sequential Quadratic Programming (TR-SSQP) method.
result Achieves high-probability first-order and second-order stationarity bounds for heavy-tailed noise.
Paper proposes a matrix optimization model for reliable Euclidean embedding from noisy data.
problem Challenges in Euclidean embedding from noisy observations containing outliers.
method Matrix optimization based embedding model to detect and remove outliers.
result The model provides high accuracy estimators and successfully identifies outliers.
The paper improves PAC-Bayes bounds for losses with finite moments.
problem Bounding generalization for losses with heavy tails and finite moments.
method Truncation method and PAC-Bayes bounds for unbounded losses with heavy tails and bounded variance.
result Bounds interpolate between slow and fast rates depending on the moment.
Develops high-probability minimax quantile bounds for statistical problems.
problem Statistical procedures often lose information about tail behavior when reduced to expectations.
method Introduces minimax quantiles, develops high-probability variants of minimax methods, and converts risk lower bounds to quantile lower bounds.
result Obtains high-probability minimax quantile lower bounds for various statistical problems.
Paper develops PAC-Bayes bounds for unknown linear systems.
problem Learning controllers for unknown stochastic linear discrete-time systems.
method PAC-Bayes framework for data-dependent high probability bounds.
result Proposes efficient learning algorithms with theoretical guarantees.
Study contextual bandits with stage-wise constraints, proving regret bounds and extending results.
problem Contextual bandits with stage-wise constraints in high probability and expectation settings.
method Upper-confidence bound algorithms for linear and non-linear reward/cost functions, extending to multiple constraints.
result Regret bounds for various settings, including non-linear reward/cost functions.
A novel post-hoc calibration method reduces neural network calibration errors.
problem Neural networks produce poorly calibrated probabilities, leading to underconfidence and overconfidence.
method Probability bounding (PB) via box-constrained softmax (BCSoftmax) function.
result Consistently reduces calibration errors on four real-world datasets.
Paper develops bounds for stochastic approximation with averaging.
problem Establish high-probability bounds for averaged stochastic approximation.
method Develops a general framework for non-asymptotic concentration bounds.
result Derives sharp bounds for averaged iterates and tightens existing results.
New bounds quantify estimation error in kernel-based system identification with unknown hyperparameters.
problem Inaccurate error bounds for kernel-based system identification with unknown hyperparameters.
method Construct a high-probability set for true hyperparameters from marginal likelihood, then find worst-case posterior covariance.
result Proposed bounds contain true model with high probability and verified in simulations.
Improved SGD bounds for machine learning models with Markovian noise.
problem Uniform high-probability bounds for SGD under PL condition with Markovian noise.
method Combining Poisson equation for Markovian noise and probabilistic induction for almost-sure bounds.
result Matching 1/k decay rate for expected suboptimality. New algorithm FLUTE achieves uniform-PAC convergence in RL with linear approx.
problem RL with linear function approximation lacks uniform-PAC guarantees.
method FLUTE algorithm with minimax value function estimator and multi-level partition scheme.
result Uniform-PAC convergence to optimal policy with high probability.
The softmax representation of probabilities for categorical variables plays a prominent role in modern machine learning with numerous applications in areas such as large scale classification, neural language modeling and recommendation systems. However, softmax estimation is very expensive for large scale inference bec…
Consider the problem of minimizing functions that are Lipschitz and strongly convex, but not necessarily differentiable. We prove that after T steps of stochastic gradient descent, the error of the final iterate is O(log(T)/T) with high probability. We also construct a function from this class for which the error …
The paper provides high-probability bounds on false discovery proportions in conformal inference.
problem Existing methods fail to provide high-probability bounds on the realized false discovery proportion.
method Constructing a high-probability envelope for the empirical distribution function of null conformal p-values by sampling from their joint distribution.
result Establishes finite-sample, distribution-free upper bounds on the FDP that hold simultaneously over all possible rejection thresholds.
Quantum probability metrics improve distribution comparison in high dimensions.
problem Challenges in comparing probability distributions, especially in high-dimensional and non-compact domains.
method Quantum probability metrics (QPMs) derived from quantum state spaces, overcoming limitations of MMD.
result QPMs offer enhanced sensitivity to subtle distributional differences in high dimensions and improve performance in generative modeling.
Improved location estimation for high-dimensional data with finite sample size.
problem Estimating the shift in high-dimensional data with limited samples.
method Smoothed estimators and bounds on subgamma vectors.
result Convergence to Cramér-Rao bound for finite sample sizes.
Paper extends tail bounds to high-dimensional random objects on Riemannian manifolds.
problem Need for tail bounds in high-dimensional data.
method Random walks on graph approximating the manifold, ensuring spectral similarity.
result Derived tensor Chernoff bound for Riemannian manifolds.
The paper offers precise bounds for averaged LSA iterates in linear systems.
problem Computing approximate solutions of linear systems with noisy observations.
method Finite-time analysis of LSA algorithms with Polyak-Ruppert averaging.
result Sharp high-probability bounds for averaged LSA iterates.
In this paper, the problem of maximizing a black-box function f:X→R is studied in the Bayesian framework with a Gaussian Process (GP) prior. In particular, a new algorithm for this problem is proposed, and high probability bounds on its simple and cumulative regret are established. The query po…
Standard results in stochastic convex optimization bound the number of samples that an algorithm needs to generate a point with small function value in expectation. More nuanced high probability guarantees are rare, and typically either rely on "light-tail" noise assumptions or exhibit worse sample complexity. In this …
New algorithm finds critical points in non-convex optimization with heavy-tailed gradients.
problem Non-convex stochastic optimization with heavy-tailed gradient estimates.
method Gradient clipping, momentum, and normalized gradient descent.
result High-probability convergence to critical points with best-known rates.
The paper optimizes distribution estimation with high probability in Kullback-Leibler divergence.
problem Estimating discrete distributions with high probability in Kullback-Leibler divergence.
method Uses online learning techniques for novel estimator construction via online-to-batch conversion.
result Optimal rate of estimation is pinned down up to a doubly logarithmic factor of K.