arXiv research

New algorithm BEAR reduces instability in off-policy Q-learning.

This paper presents the hierarchical generalized linear model (HGLM) for loss reserving in a non-life insurance company. Because in this case the error of prediction is expressed by a complex analytical formula, the error bootstrap estimator is proposed instead. Moreover, the bootstrap procedure is used to obtain full …

We quantify uncertainty in Oja's algorithm's leading eigenvector estimation.

Paper develops a bootstrap method for estimating sketched SVD errors.

A fast bootstrap method estimates cross-validation standard error.

Blockwise bootstrap improves ASR performance testing for correlated data.

Paper develops methods for statistical inference with SGD in nonconvex optimization.

In this paper, we detail the main simulation methods used in practice to measure one-year reserve risk, and describe the bootstrap method providing an empirical distribution of the Claims Development Result (CDR) whose variance is identical to the closed-form expression of the prediction error proposed by Wüthrich et a…

This paper presents an alternative approach to p-values in regression settings. This approach, whose origins can be traced to machine learning, is based on the leave-one-out bootstrap for prediction error. In machine learning this is called the out-of-bag (OOB) error. To obtain the OOB error for a model, one draws a bo…

Private statistical inference methods improve confidence interval lengths.

Robust forecast framework reduces distribution error by 63%.

We consider the performance of the bootstrap in high-dimensions for the setting of linear regression, where $p<n$ but $p/n$ is not close to zero. We consider ordinary least-squares as well as robust regression methods and adopt a minimalist performance requirement: can the bootstrap give us good confidence intervals fo…

Paper improves bootstrapping for off-policy reinforcement learning inference.

Graphical lasso models ASR utterance dependencies for consistent WER estimation.

Proposes a new test for validating multivariate dynamic regression models.

This paper develops a bootstrap method to estimate errors in Random Fourier Features.

The paper tackles model misspecification in reinforcement learning through a bootstrapped neural network and error correction.

Estimates algorithmic variance for bagging and random forests using bootstrap.

New algorithm speeds up causal inference for large data.

New methods improve anomaly detection with reduced false positives.

A new estimator combines bootstrapping and rollout methods in RL.

Confidence measures for the generalization error are crucial when small training samples are used to construct classifiers. A common approach is to estimate the generalization error by resampling and then assume the resampled estimator follows a known distribution to form a confidence set [Kohavi 1995, Martin 1996,Yang…

The paper studies how to use AI-generated labels in econometrics to avoid bias.

Paper develops Gaussian approximations and bootstrap for federated LSA with trade-off bounds.

Improved pricing of vanilla options using modified Adams method and sinh-acceleration.

Paper improves CLT and bootstrap approximations for LSA with decreasing step size.

Thompson sampling provides a solution to bandit problems in which new observations are allocated to arms with the posterior probability that an arm is optimal. While sometimes easy to implement and asymptotically optimal, Thompson sampling can be computationally demanding in large scale bandit problems, and its perform…

The intention of this paper is to estimate a Bayesian distribution-free chain ladder (DFCL) model using approximate Bayesian computation (ABC) methodology. We demonstrate how to estimate quantities of interest in claims reserving and compare the estimates to those obtained from classical and credibility approaches. In …

Over the course of the past decade, a variety of randomized algorithms have been proposed for computing approximate least-squares (LS) solutions in large-scale settings. A longstanding practical issue is that, for any given input, the user rarely knows the actual error of an approximate solution (relative to the exact …

New methods improve insurance data quality for catastrophic events.

Method captures shared information across many views robustly.

Non-parametric bootstrap improves robust portfolio and trading strategy optimization.

Cross-validation estimates model performance on unseen data, not training data.

We address the problem of Bayesian structure learning for domains with hundreds of variables by employing non-parametric bootstrap, recursively. We propose a method that covers both model averaging and model selection in the same framework. The proposed method deals with the main weakness of constraint-based learning--…

The paper provides rigorous guarantees for m-out-of-n bootstrap estimators of sample quantiles.

A new method boosts exploration in bandit algorithms, reducing regret.

Enhanced TSFMs improve time series forecasting accuracy and reliability.

We study the variability of predictions made by bagged learners and random forests, and show how to estimate standard errors for these methods. Our work builds on variance estimates for bagging proposed by Efron (1992, 2012) that are based on the jackknife and the infinitesimal jackknife (IJ). In practice, bagged predi…

Study evaluates 13 CI methods for generalization error across 19 problems.

Novel method estimates CV-based classifier performance variance.

A new method for approximating CV and bootstrap with higher-order infinitesimal jackknife.

Optimizes a small set of centroid points to approximate bootstrap distribution.

New methods improve uncertainty in machine learning predictions for asset returns.

In this paper we tackle the problem of estimating the power-law tail exponent of income distributions by using the Hill's estimator. A subsample semi-parametric bootstrap procedure minimising the mean squared error is used to choose the power-law cutoff value optimally. This technique is applied to personal income data…

AR-Sieve Bootstrap improves Random Forest time series prediction accuracy.

New bootstraps improve speed and accuracy for graph count functionals.

A new method reduces bootstrap simulation cost and improves accuracy.

Three bootstrap tests compare categorical time series generating processes.