An Introduction to the Bootstrap

2007

Subfields: Statistics and Probability

Citations in the corpus

• Efficient Large Scale Language Modeling with Mixtures of Experts (2021, background): “05) difference between best and worst scores based on bootstrap test (Noreen, 1989; Efron and Tibshirani, 1994).”
• Addressing Semantic Drift in Question Generation for Semi-Supervised Question Answering (2019, methodology): “The significance tests in this paper are conducted following the bootstrap test setup (Efron and Tibshirani, 1994).”
• A-NICE-MC: Adversarial Training for MCMC (2017, methodology): “In order to retrieve samples from pd and train our model, we use a bootstrap process [33] where the quality of samples used for adversarial training should increase over time.”
• Do Language Models Have Beliefs? Methods for Detecting, Updating, and Visualizing Model Beliefs (2021, methodology): “We obtain 95% confidence intervals and perform hypothesis tests via block bootstrap, resampling model seeds and data points (Efron and Tibshirani, 1994).”
• Deep Reinforcement Learning that Matters (2017, methodology): “Confidence Bounds The sample bootstrap has been a popular method to gain insight into a population distribution from a smaller sample (Efron and Tibshirani 1994).”
• Show Your Work: Improved Reporting of Experimental Results (2019, methodology): “The bootstrap (Efron and Tibshirani, 1994) is a general method which can be used to estimate statistics that do not have a closed form.”
• Multi-Reward Reinforced Summarization with Saliency and Entailment (2018, neutral): “4 Our baseline is statistically equal to the paper-reported scores of See et al. (2017) (see Table 1) on ROUGE-1, ROUGE-2, based on the bootstrap test (Efron and Tibshirani, 1994).”
• PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examples (2017, methodology): “Combined with permutation test (Efron & Tibshirani, 1994), we can provide a uncertainty value for each input about whether it comes from the training distribution or not.”
• Security Evaluation of Pattern Classifiers under Attack (2014, methodology): “Accordingly, we can sample with replacement from D TR [42].”
• Robust multimodal face and fingerprint fusion in the presence of spoofing attacks (2016, methodology): “The bootstrap method facilitates determining the probability distribution of the data without using the Central Limit Theorem [34].”
• Continual and Multi-Task Architecture Search (2019, positive): “01, and statistically equal on RTE and WNLI (where the validations sets are very small), based on the bootstrap test (Noreen, 1989; Efron and Tibshirani, 1994) with 100K samples.”
• Data-Efficient Off-Policy Policy Evaluation for Reinforcement Learning (2016, methodology): “∈Z We use both the percentile bootstrap confidence interval (Efron & Tibshirani, 1993) and Chernoff-Hoeffding’s inequality—whichever is tighter—for CI in our experiments.”
• Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models (2018, background): “model, no additional (hyper-)parameters need be tuned, whilst still providing reasonable uncertainty estimates [47], [36].”
• Continual World: A Robotic Benchmark For Continual Reinforcement Learning (2021, methodology): “[14] Bradley Efron and Robert J Tibshirani.”
• Continual Few-Shot Learning for Text Classification (2021, background): “Bold numbers are statistically significantly better than the rest based on bootstrap test (Efron and Tibshirani, 1994).”
• Uncertainty-Aware Reinforcement Learning for Collision Avoidance (2017, methodology): “Bootstrapping : Bootstrapping [5, 6] is a simple and effective method of estimating model uncertainty using resampling that can be used with any discriminatively trained model.”
• Deep Exploration via Bootstrapped DQN (2016, neutral): “The bootstrap princple is to approximate a population distribution by a sample distribution [6].”
• Informing sequential clinical decision-making through reinforcement learning: an empirical study (2010, methodology): “Resampling methods such as the bootstrap are often used in the clinical trial setting to provide both approximate variance estimates and confidence intervals (Efron 1979; Efron and Tibshirani 1993).”
• PREDICTIVE LEARNING VIA RULE ENSEMBLES (2008, methodology): “Here we present a variant of the parametric bootstrap [Efron and Tibshirani (1993)] that can be used to derive a reference (null) distribution for any of the interaction test statistics presented in Section 8.”