An official website of the United States government
A<sc>bstract</sc> Signal-agnostic data exploration based on machine learning could unveil very subtle statistical deviations of collider data from the expected Standard Model of particle physics. The beneficial impact of a large training sample on machine learning solutions motivates the exploration of increasingly large and inclusive samples of acquired data with resource efficient computational methods. In this work we consider the New Physics Learning Machine (NPLM), a multivariate goodness-of-fit test built on the Neyman-Pearson maximum-likelihood-ratio construction, and we address the problem of testing large size samples under computational and storage resource constraints. We propose to perform parallel NPLM routines over batches of the data, and to combine them by locally aggregating over the data-to-reference density ratios learnt by each batch. The resulting data hypothesis defining the likelihood-ratio test is thus shared over the batches, and complies with the assumption that the expected rate of new physical processes is time invariant. We show that this method outperforms the simple sum of the independent tests run over the batches, and can recover, or even surpass, the sensitivity of the single test run over the full data. Beside the significant advantage for the offline application of NPLM to large size samples, the proposed approach offers new prospects toward the use of NPLM to construct anomaly-aware summary statistics in quasi-online data streaming scenarios.
more »
« less
Multiple testing for signal-agnostic searches for new physics with machine learning
Abstract In this work, we address the question of how to enhance signal-agnostic searches by leveraging multiple testing strategies. Specifically, we consider hypothesis tests relying on machine learning, where model selection can introduce a bias towards specific families of new physics signals. Focusing on the New Physics Learning Machine, a methodology to perform a signal-agnostic likelihood-ratio test, we explore a number of approaches to multiple testing, such as combiningp-values and aggregating test statistics. Our findings show that it is beneficial to combine different tests, characterised by distinct choices of hyperparameters, and that performances comparable to the best available test are generally achieved, while also providing a more uniform response to various types of anomalies. This study proposes a methodology that is valid beyond machine learning approaches and could in principle be applied to a larger class model-agnostic analyses based on hypothesis testing.
more »
« less
- Award ID(s):
- 2019786
- PAR ID:
- 10564784
- Publisher / Repository:
- Springer Science + Business Media
- Date Published:
- Journal Name:
- The European Physical Journal C
- Volume:
- 85
- Issue:
- 1
- ISSN:
- 1434-6052
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Researchers theorize that many real-world networks exhibit community structure where within-community edges are more likely than between-community edges. While numerous methods exist to cluster nodes into different communities, less work has addressed this question: given some network, does it exhibitstatistically meaningfulcommunity structure? We answer this question in a principled manner by framing it as a statistical hypothesis test in terms of a general and model-agnostic community structure parameter. Leveraging this parameter, we propose a simple and interpretable test statistic used to formulate two separate hypothesis testing frameworks. The first is an asymptotic test against a baseline value of the parameter while the second tests against a baseline model using bootstrap-based thresholds. We prove theoretical properties of these tests and demonstrate how the proposed method yields rich insights into real-world datasets.more » « less
-
Abstract We develop alternative families of Bayes factors for use in hypothesis tests as alternatives to the popular default Bayes factors. The alternative Bayes factors are derived for the statistical analyses most commonly used in psychological research – one-sample and two-samplet tests, regression, and ANOVA analyses. They possess the same desirable theoretical and practical properties as the default Bayes factors and satisfy additional theoretical desiderata while mitigating against two features of the default priors that we consider implausible. They can be conveniently computed via an R package that we provide. Furthermore, hypothesis tests based on Bayes factors and those based on significance tests are juxtaposed. This discussion leads to the insight that default Bayes factors as well as the alternative Bayes factors are equivalent to test-statistic-based Bayes factors as proposed by Johnson.Journal of the Royal Statistical Society Series B: Statistical Methodology,67, 689–701. (2005). We highlight test-statistic-based Bayes factors as a general approach to Bayes-factor computation that is applicable to many hypothesis-testing problems for which an effect-size measure has been proposed and for which test power can be computed.more » « less
-
Although measuring held-out accuracy has been the primary approach to evaluate generalization, it often overestimates the performance of NLP models, while alternative approaches for evaluating models either focus on individual tasks or on specific behaviors. Inspired by principles of behavioral testing in software engineering, we introduce CheckList, a task-agnostic methodology for testing NLP models. CheckList includes a matrix of general linguistic capabilities and test types that facilitate comprehensive test ideation, as well as a software tool to generate a large and diverse number of test cases quickly. We illustrate the utility of CheckList with tests for three tasks, identifying critical failures in both commercial and state-of-art models. In a user study, a team responsible for a commercial sentiment analysis model found new and actionable bugs in an extensively tested model. In another user study, NLP practitioners with CheckList created twice as many tests, and found almost three times as many bugs as users without it.more » « less
-
Abstract Conjoint analysis is a popular experimental design used to measure multidimensional preferences. Many researchers focus on estimating the average marginal effects of each factor while averaging over the other factors. Although this allows for straightforward design-based estimation, the results critically depend on the ways in which factors interact with one another. An alternative model-based approach can compute various quantities of interest, but requires correct model specifications, a challenging task for conjoint analysis with many factors. We propose a new hypothesis testing approach based on the conditional randomization test (CRT) to answer the most fundamental question of conjoint analysis: Does a factor of interest matterin any waygiven the other factors? Although it only provides a formal test of these binary questions, the CRT is solely based on the randomization of factors, and hence requires no modeling assumption. This means that the CRT can provide a powerful and assumption-free statistical test by enabling the use of any test statistic, including those based on complex machine learning algorithms. We also show how to test commonly used regularity assumptions. Finally, we apply the proposed methodology to conjoint analysis of immigration preferences. An open-source software package is available for implementing the proposed methodology. The proposed methodology is implemented via an open-source software R packageCRTConjoint, available through the Comprehensive R Archive Networkhttps://cran.r-project.org/web/packages/CRTConjoint/index.html.more » « less
