In this paper, we give a new sharp generalization bound of lp-MKL which is a generalized framework of multiple kernel learning (MKL) and imposes lp-mixed-norm regularization instead of l1-mixed-norm regularization. We utilize localization techniques to obtain the sharp learning rate. The bound is characterized by the d…
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
168,657 papers · 148 categories
Trend · papers per month
48 results for “mixing rates”
The literature on statistical learning for time series assumes the asymptotic independence or ``mixing' of the data-generating process. These mixing assumptions are never tested, nor are there methods for estimating mixing rates from data. We give an estimator for the -mixing rate based on a single stationary sample…
This paper investigates the supervised learning problem with observations drawn from certain general stationary stochastic processes. Here by \emph{general}, we mean that many stationary stochastic processes can be included. We show that when the stochastic processes satisfy a generalized Bernstein-type inequality, a u…
Mixed likelihood GPs improve model performance in human-in-the-loop experiments.
problem Lack of auxiliary information in traditional GPs for human responses.
method Propose mixed likelihood variational GPs to leverage auxiliary information.
result Modeling performance improvements across diverse applications.
Study online learning in RKHS with dependent processes, focusing on \(β\)- and \(φ\)-mixing.
problem Online learning in RKHS with dependent data.
method Online regularized learning algorithm in RKHS, analyzing \(β\)- and \(φ\)-mixing sequences.
result Probabilistic upper bounds and convergence rates for mixing coefficients.
This paper focuses on learning rate analysis of distributed kernel ridge regression for strong mixing sequences. Using a recently developed integral operator approach and a classical covariance inequality for Banach-valued strong mixing sequences, we succeed in deriving optimal learning rate for distributed kernel ridg…
This work improves mixing rates for Bayesian CART, a key component of BART.
problem Understanding and improving mixing rates for Bayesian inference with MCMC.
method Derived upper bounds on mixing times, provided sufficient conditions for polynomial mixing, and proposed Twiggy Bayesian CART.
result Twiggy Bayesian CART achieves polynomial mixing without assuming signal connectivity.
Paper analyzes Nyström regularization for time series forecasting with sequential sub-sampling.
problem Learning rate analysis of Nyström regularization for -mixing time series.
method Banach-valued Bernstein inequality and integral operator approach for -mixing sequences.
result Almost optimal learning rates for Nyström regularization with sequential sub-sampling.
Sharp rates found for learning with dependent data, avoiding sample size deflation.
problem Learning with dependent data and square loss.
method Combining weak sub-Gaussian class and mixed tail generic chaining.
result Achieves a rate that only depends on class complexity and second order statistics.
Elo ratings learn model parameters quickly using Markov chains.
problem Ranking players in online settings.
method Bradley--Terry--Luce model and Markov chain theory.
result Elo learns model parameters at a competitive rate.
Approximation and learning of anisotropic and mixed smooth functions by deep ReLU neural networksstat.ML
Deep ReLU networks can approximate and learn smooth functions efficiently.
problem Efficiently approximating and learning smooth functions using deep ReLU neural networks.
method Extending recent results to anisotropic and mixed smooth function classes, establishing approximation rates.
result Deep ReLU networks achieve minimax optimal rates up to logarithmic factors for various smooth function classes.
In this paper we address the following question: Can we approximately sample from a Bayesian posterior distribution if we are only allowed to touch a small mini-batch of data-items for every sample we generate?. An algorithm based on the Langevin equation with stochastic gradients (SGLD) was previously proposed to solv…
Optimizes pension mix of PAYGO, EET, and individual savings.
problem Balancing PAYGO, EET, and individual savings in funded pension schemes.
method Solves a Nash equilibrium between pension participants and government, considering age-dependent preferences and optimal asset allocation.
result Identifies critical ages and optimal contribution rates for maximizing overall utility.
TD(0) with Polyak-Ruppert averaging achieves robust and fast convergence rates
problem TD(0) learning under Markovian sampling
method Polyak-Ruppert averaging with a single stepsize
result Simultaneous high-probability convergence guarantees for TD(0) iterates and PR average
The paper tackles high-dimensional mixed linear regression with unknown parameters and proposes methods for estimation, confidence intervals, and hypothesis testing.
problem High-dimensional mixed linear regression with unknown parameters and covariance structure.
method Iterative high-dimensional EM algorithm for estimating regression vectors, debiased estimators for individual coordinates, and large-scale multiple testing procedure.
result Asymptotic normality of debiased estimators and FDR control for hypothesis testing.
Improves decentralized learning by optimizing graph mixing for data heterogeneity.
problem Data heterogeneity impacts convergence in decentralized learning, but existing methods ignore this.
method Characterized and quantified the relationship between graph mixing and data heterogeneity. Proposed an optimization approach to improve convergence.
result Our approach leads to improved test performance across various tasks.
Mixed-SCORE+ improves community detection in weak signal networks.
problem Detecting communities in weak signal networks.
method Proposes Mixed-SCORE+ combining properties of Mixed-SCORE and SCORE+.
result Significantly improves detection error rates on Polblogs and weak signal networks.
AM converges super-linearly for solving mixed linear regression problems.
problem Learning linear regressors from unlabeled observations in multiple linear regression models.
method Alternating Minimization (AM) algorithm, which alternates between label estimation and regression solving.
result AM converges super-linearly in certain parameter regimes, requiring only O(log log(1/ε)) iterations to achieve an error of ε.
New RL method MAC improves performance in sparse reward settings.
problem Slow mixing in large state spaces or sparse rewards.
method Multi-level Monte Carlo Actor-Critic (MAC) algorithm.
result Achieves convergence rate comparable to state-of-the-art AC algorithms.
Beyond the Independence Assumption: Finite-Sample Guarantees for Deep Q-Learning under -Mixingstat.ML
Study shows DQN's performance degrades with temporal dependence in data.
problem Temporal dependence in replayed data affects DQN's performance.
method Modelled -mixing data, derived risk bounds, and empirical validation.
result Temporal dependence leads to a degradation in DQN's performance rate.
Temporal Difference Learning analysis under non-i.i.d. data and nonlinear approximation.
problem Finite-sample behavior of TD(0) under non-i.i.d. data and nonlinear approximation.
method High-probability, finite-sample analysis of vanilla TD(0) on polynomially mixing Markov data, assuming Holder continuity and bounded generalized gradients.
result Bounds on the convergence rate of TD(0) with high probability, matching known i.i.d. rates and holding even with nonstationary initialization.
Study fast learning rates for square loss in dependent data with hypercontractivity condition.
problem Learning from dependent data with fast rates matching independent data.
method Martingale difference noise, trajectory hypercontractivity condition, least-squares estimator.
result Excess risk bound matches iid rate after burn-in time, independent of mixing-time.
We establish exponential mixing for the geodesic flow of an incomplete, negatively curved surface with cusp-like singularities of a prescribed order. As a consequence, we obtain that the Weil-Petersson flows for the moduli spaces and are expon…
Estimates mixing coefficients of geometrically ergodic Markov processes from a single sample path.
problem Estimating mixing coefficients of geometrically ergodic Markov processes.
method Proposes methods to estimate -mixing coefficients from a single sample path under standard smoothness conditions.
result Obtains a rate of convergence of order \(\mathcal{O}(\log(n) n^{-[s]/(2[s]+2)})\) for the expected error of the estimator.
New method adapts to unknown mixing time in stochastic optimization.
problem Optimizing with Markovian data where mixing time is unknown.
method Combines MLMC gradient estimation with adaptive learning.
result Achieves optimal convergence rate for convex problems.
Paper proposes deep neural networks for nonparametric regression from dependent data.
problem Nonparametric regression from strongly mixing observations.
method Minimum error entropy principle applied to deep neural networks.
result Deep neural networks achieve minimax optimal convergence rates for Gaussian errors.
Develops a bi-variate stochastic framework to model mortality and interest rates with long-range dependence.
problem Captures long-range dependence and instantaneous correlation in mortality and interest rates.
method Mixed fractional Brownian motions, analytical solutions, risk-neutral measure, sequential parameter estimation.
result Explicit pricing of zero-coupon bonds and extreme mortality bonds, practical implications for pricing and risk management.
On the Minimax Optimality of the EM Algorithm for Learning Two-Component Mixed Linear Regressionstat.ML
EM algorithm achieves optimal sample complexity for learning two-component mixed linear regression.
problem Learning two-component mixed linear regression under varying signal-to-noise ratios.
method Analysis of EM algorithm convergence rates under different SNR regimes.
result EM algorithm achieves minimax optimal sample complexity in all SNR regimes.
Nearly Maximally Predictive Features and Their Dimensionscond-mat.stat-mech
Scientific explanation often requires inferring maximally predictive features from a given data set. Unfortunately, the collection of minimal maximally predictive features for most stochastic processes is uncountably infinite. In such cases, one compromises and instead seeks nearly maximally predictive features. Here, …
Develops a contraction framework for MCMC mixing rates.
problem Proving mixing-time bounds for MCMC algorithms.
method Global and local contraction coefficients under -divergence.
result Explicit global contraction coefficients for Gaussian smoothing.
Embedding representations power machine intelligence in many applications, including recommendation systems, but they are space intensive -- potentially occupying hundreds of gigabytes in large-scale settings. To help manage this outsized memory consumption, we explore mixed dimension embeddings, an embedding layer arc…
Paper develops a novel approach for classifying high-dimensional mixed data.
problem Handling datasets with both categorical and continuous variables of high dimensions.
method Location model with Gaussian conditional distributions, kernel smoothing for bandwidth choice, penalized likelihood estimation.
result Competitive performance of the proposed classifier demonstrated through simulations and real data.
Accelerates sampling from Gibbs distributions using ARWP method.
problem Sampling from Gibbs distributions efficiently.
method ARWP method, combining Nesterov acceleration and regularized Wasserstein proximal.
result ARWP exhibits higher contraction rate and faster tail exploration.
About the Justification of Experience Rating: Bonus Malus System and a new Poisson Mixture Modelq-fin.RM
The claim experience of the past is a very important information to calculate the fair price of an insurance contract. In a lot of European countries for instance the prices for motor car insurance depend on the number of claims the driver has reported to the insurance company during the last years. Classically these p…
This paper studies convergence behavior of latent mixing measures that arise in finite and infinite mixture models, using transportation distances (i.e., Wasserstein metrics). The relationship between Wasserstein distances on the space of mixing measures and f-divergence functionals such as Hellinger and Kullback-Leibl…
Ballistic Convergence in Hit-and-Run Monte Carlo and a Coordinate-free Randomized Kaczmarz Algorithmmath.PR
Paper analyzes Hit-and-Run's convergence rates and applies similar methods to randomized Kaczmarz.
problem Quantifying advantages of Hit-and-Run's coordinate-free property.
method Sharp estimates via coupling methods and mixing time bounds.
result Ballistic and superdiffusive convergence rates in certain settings.
DiMMSB models directed mixed membership networks, identifying distinct community structures.
problem Modeling directed mixed membership networks with distinct community structures.
method Directed Mixed Membership Stochastic Blockmodel (DiMMSB) with DiSP algorithm.
result DiSP algorithm is asymptotically consistent and outperforms competitors.
New framework relaxes independence assumption for graph-mixing dependencies.
problem Tackles limitations of existing generalization results for graph-mixing dependencies.
method Proposes a framework where dependencies decay with graph distance, derives generalization bounds leveraging online-to-PAC framework.
result Derives high-probability generalization guarantees that depend on mixing rate and graph's chromatic number.
MALA mixes optimally in κ√d steps for log-concave sampling.
problem Sampling from log-concave distributions efficiently.
method Metropolis-Adjusted Langevin Algorithm (MALA) with warm start.
result Optimal minimax mixing time of κ√d iterations for log-concave distributions.
The paper tackles long-context linear system identification with improved sample complexity bounds.
problem Identifying dynamical systems with long dependencies over fixed context windows.
method Established sample complexity bounds for systems with linear dependencies over a context window of length p.
result The learning process is not hindered by slow mixing properties in extended context windows.
We reconsider the training objective of Generative Adversarial Networks (GANs) from the mixed Nash Equilibria (NE) perspective. Inspired by the classical prox methods, we develop a novel algorithmic framework for GANs via an infinite-dimensional two-player game and prove rigorous convergence rates to the mixed NE, reso…
We study a special case of the problem of statistical learning without the i.i.d. assumption. Specifically, we suppose a learning method is presented with a sequence of data points, and required to make a prediction (e.g., a classification) for each one, and can then observe the loss incurred by this prediction. We go …
Deep learning has shown high performances in various types of tasks from visual recognition to natural language processing, which indicates superior flexibility and adaptivity of deep learning. To understand this phenomenon theoretically, we develop a new approximation and estimation error analysis of deep learning wit…
The paper analyzes convergence rates of Langevin dynamics and Proximal Sampler using -divergence.
problem Analyzing convergence rates of Langevin dynamics and Proximal Sampler.
method Extending mixing time analyses to -divergence, using strong data processing inequalities.
result Convergence of -divergence to 0 exponentially fast along Unadjusted Langevin Algorithm and Proximal Sampler.
In this paper, the valuation of European and path-dependent options in foreign exchange (FX) markets is considered when the currency exchange rate evolves according to the Heston model combined with the Cox-Ingersoll-Ross dynamics for the stochastic domestic and foreign short interest rates. The mixed Monte Carlo/PDE m…
New bandit algorithm for non-i.i.d. noise, improving standard rates.
problem Linear stochastic bandit with non-i.i.d. observation noise.
method Developed new confidence sequences and an algorithm based on optimism in uncertainty.
result Regret bounds for the new algorithm, showing recovery of standard rates up to a factor of the mixing time.
This paper mixes constant sum and constant product market makers to improve their features.
problem Improving the balance between stable exchange rates and liquidity in automated market makers.
method Mixing and designing new methods for AMMs with specific features.
result Demonstrates new tools for creating markets with desired characteristics.
A new method combines machine learning with mixed-effects models for better repeated measurement analysis.
problem Inference of linear coefficients in partially linear mixed-effects models with complex interactions and high-dimensional variables.
method Double machine learning approach to estimate nonparametrically nonlinear variables, then use standard linear mixed-effects techniques to estimate the linear coefficient.
result The estimated fixed effects coefficient converges at the parametric rate and is semiparametrically efficient.