Simple methods combine statistical tests for out-of-distribution detection.
problem Detecting data points not following the training distribution.
method Combining classical parametric tests (Rao's score test) and a typicality test.
result Combining Fisher's method of test statistics improves out-of-distribution detection accuracy.
Unified score and distance-based GoF tests for model adequacy.
problem Difficulty in extending score-based GoF tests to nonparametric alternatives.
method Introducing semiparametric kernelized Stein discrepancy (SKSD) test.
result SKSD test is computationally efficient and universally consistent.
Model-X test detects conditional independence in streaming data.
problem Detecting conditional independence in data streams with arbitrary dependency.
method Sequential testing inspired by model-X and testing by betting.
result Significantly reduces type-I error rate and enhances data efficiency.
Proposes a new test for validating multivariate dynamic regression models.
problem Inadequate exogeneity conditions for conventional model specification tests in dynamic systems.
method Develops a generalized Durbin estimator for multiple-equation systems with dynamic dependencies, and constructs Wald tests.
result Bootstrap-based Wald tests improve finite-sample size control and validate the null hypothesis in multifactor models.
Conditional independence tests (CI tests) have received special attention lately in Machine Learning and Computational Intelligence related literature as an important indicator of the relationship among the variables used by their models. In the field of Probabilistic Graphical Models (PGM)--which includes Bayesian Net…
Unified CI test for categorical and ordinal data maintains power in high dimensions.
problem Rapid degradation of statistical power in existing CI tests for high-dimensional conditioning variables.
method Unified CI test for categorical and ordinal data, maintaining reasonable calibration and power in high dimensions.
result Our test outperforms existing baselines in model testing and structure learning for dense directed graphical models.
Active testing reduces label costs for efficient model evaluation.
problem Real-world applications require expensive test labels, disconnecting from existing model evaluation methods.
method Derives acquisition strategies to select test points efficiently, addressing label bias and variance.
result Active testing improves model evaluation efficiency without sacrificing accuracy.
This paper explores the information-theoretic limitations of graph property testing in zero-field Ising models. Instead of learning the entire graph structure, sometimes testing a basic graph property such as connectivity, cycle presence or maximum clique size is a more relevant and attainable objective. Since property…
Simpler, faster algorithm for uniformity testing in the shuffle model.
problem Testing uniformity of data in the shuffle model with privacy constraints.
method Simplified analysis and use of privacy amplification via shuffling.
result An algorithm with the same guarantees but simpler and more streamlined.
Paper improves power of conditional randomization tests.
problem Improving power of conditional randomization tests.
method Introducing a new cost function to maximize test statistic power.
result Consistently increases the number of correct discoveries.
New Bayesian optimization models for efficient material screening.
problem Efficiently screening materials with expensive and cheap tests.
method Flexible multi-test Bayesian optimization models with complex relationships.
result Demonstrated power on synthetic and real data.
Risk monitoring detects when TTA models degrade at test time.
problem Detecting when TTA models degrade at test time.
method Extended risk monitoring tools based on sequential testing with confidence sequences.
result Demonstrated effectiveness of TTA monitoring framework across various datasets and methods.
We propose a novel adaptive test of goodness-of-fit, with computational cost linear in the number of samples. We learn the test features that best indicate the differences between observed samples and a reference model, by minimizing the false negative rate. These features are constructed via Stein's method, meaning th…
The repeated community-wide reuse of test sets in popular benchmark problems raises doubts about the credibility of reported test-error rates. Verifying whether a learned model is overfitted to a test set is challenging as independent test sets drawn from the same data distribution are usually unavailable, while other …
This work identifies redundant tests in conditional-independence-based discovery that can improve graphical model accuracy.
problem Reliability and sensitivity of conditional-independence-based discovery algorithms.
method Analysis of redundant tests and their impact on error detection and correction.
result Redundant tests can improve graphical model accuracy but not all are beneficial.
The increasing inclusion of Deep Learning (DL) models in safety-critical systems such as autonomous vehicles have led to the development of multiple model-based DL testing techniques. One common denominator of these testing techniques is the automated generation of test cases, e.g., new inputs transformed from the orig…
New adaptive test for NPIV models controls size and has superior power.
problem Testing inequality and equality restrictions in nonparametric IV models.
method Adaptive hypothesis test based on modified leave-one-out sample quadratic distance.
result Adaptive test attains the adaptive minimax rate of testing in L2. This paper applies combinatorial testing to machine learning for robust model performance.
problem Identifying robust machine learning models using test and training sets.
method Adapting combinatorial interaction testing for machine learning, focusing on simple features.
result Combinatorial coverage can enhance model performance and robustness.
Proposes a modified Morgan-Pitman test for evaluating variances in machine learning models.
problem Limited ability to account for sampling variability in model selection.
method Enhances the classic Morgan-Pitman test for robustness in non-linear models with heavy-tailed distributions or outliers.
result Demonstrates the test's effectiveness and practical utility in model evaluation and selection.
New methods for CI testing under model misspecification.
problem Challenges in CI testing with misspecified models.
method Proposes new approximations and upper bounds for testing errors of regression-based CI tests.
result Introduces the Rao-Blackwellized Predictor Test (RBPT) robust against misspecified inductive biases.
[Context:] Model-based testing is an instrument for automated generation of test cases. It requires identifying requirements in documents, understanding them syntactically and semantically, and then translating them into a test model. One light-weight language for these test models are Cause-Effect-Graphs (CEG) that ca…
Test log-likelihood comparisons can be misleading.
problem Misinterpretation of test log-likelihood in model comparison.
method Simple examples of model comparison and forecast accuracy.
result Test log-likelihood does not always correlate with model accuracy.
Paper proposes a robust test for high-dimensional models with large covariates and instruments.
problem Testing high-dimensional linear instrumental variable models with large covariates and instruments.
method Introduces a test based on the maximum norm of multiple parameters and a power-enhanced test.
result The proposed test is robust to heteroskedastic errors and has higher power than existing tests.
We propose a kernel-based nonparametric test of relative goodness of fit, where the goal is to compare two models, both of which may have unobserved latent variables, such that the marginal distribution of the observed variables is intractable. The proposed test generalizes the recently proposed kernel Stein discrepanc…
Hypothesis testing in singular models is fundamentally about identifiable vs. non-identifiable parameters.
problem Testing in singular models is inherently problematic due to non-identifiability and degeneracy of Fisher information.
method Formalized the overlap obstruction and showed that hypotheses over non-identifiable parameters are untestable, while those over identifiable parameters reduce to classical testing.
result Hypotheses over non-identifiable parameters are untestable, while those over identifiable parameters reduce to classical testing.
Tests if vertices in graphs have the same latent positions.
problem Testing equality of latent positions in random graphs.
method Empirical Mahalanobis distances from spectral embeddings.
result Test statistics follow chi-square distributions under null and local alternatives.
Tests for equivariance in non-parametric regression models.
problem Detecting false assumptions of symmetry in regression models.
method Develops tests for G-equivariance independent of the model. result Confidence in using equivariant models when symmetry is unknown.
We propose a new probabilistic graphical model that jointly models the difficulties of questions, the abilities of participants and the correct answers to questions in aptitude testing and crowdsourcing settings. We devise an active learning/adaptive testing scheme based on a greedy minimization of expected model entro…
A new test for IRG models using KSD for small networks.
problem Testing goodness-of-fit for inhomogeneous random graph models.
method Kernelised Stein Discrepancy (KSD) test for IRG models.
result The test is effective for small networks and provides theoretical guarantees.
A method uses neural networks to approximate sampling distributions of test statistics.
problem Accurate modeling of p-value functions or cdfs for correct confidence set coverage.
method Uses neural networks to model the cdf of test statistics, approximating sampling distributions.
result Neural network approximations of sampling distributions are effective and simple.
Proposes a new model for testing causal structural priors and synthesizing data.
problem Testing and synthesizing causal structural priors using nonparametric knowledge and neural networks.
method Causal Structural Hypothesis Testing (C-SHT) and Causal Structural Variational Hypothesis Testing (C-SVHT) using deep neural networks.
result Demonstrates out-of-distribution generalization error as a proxy for causal structural prior hypothesis testing.
Deep-learning method improves hypothesis testing for independence.
problem Improving hypothesis testing for independence using deep learning.
method Proposes deep-testing, a novel procedure that uses a deep neural network to distinguish between data generated under and outside a given statistical model.
result Deep-testing achieves the highest overall power against nineteen competing methods across various dependence structures.
Latent block models are used for probabilistic biclustering, which is shown to be an effective method for analyzing various relational data sets. However, there has been no statistical test method for determining the row and column cluster numbers of latent block models. Recent studies have constructed statistical-test…
Framework for online hypothesis testing across various data types.
problem Testing various nonparametric hypotheses in data streams.
method Unified framework using operators on data distributions, leveraging ML models.
result Efficient, adaptive, and error-controlled sequential tests.
A new test assesses how well observed networks fit a specified ERGM model.
problem Testing the goodness of fit for ERGMs with a single network observation.
method Kernel Stein discrepancy combined with a discrete Stein operator for ERGMs, Monte Carlo simulation.
result The test provides theoretical and practical support for assessing ERGM fit.
Active testing for large language models is made more efficient and accurate.
problem Efficient evaluation of large language models with limited labels.
method Cost-saving measures and in-context learning for constructing a surrogate model.
result Significantly more accurate evaluations of LLM performance compared to random data acquisition.
New framework compares credal sets for hypothesis testing with epistemic uncertainty.
problem Comparing distributions with partial ignorance and epistemic uncertainty.
method Credal two-sample testing framework for convex sets of probability measures.
result Direct integration of epistemic uncertainty in hypothesis testing.
A test for comparing networks using stochastic block models.
problem Determining if two network datasets come from the same model.
method Adopting stochastic block models, the study introduces an efficient algorithm to match estimated network parameters and develops a powerful test.
result The test is consistent and asymptotically follows a chi-squared distribution.
The paper tests properties of trees in graphical models using covariance queries.
problem Testing properties of trees in graphical models.
method Covariance queries model, randomized tests for tree properties.
result Efficient testing of global tree properties using sub-quadratic number of queries.
Paper tests DPPs for diversity models, distinguishing them from other distributions.
problem Testing whether a given distribution is a Determinantal Point Process (DPP) or far from any DPP.
method Proposes the first algorithm for DPP testing and establishes a lower bound on sample complexity.
result Establishes a matching lower bound on the sample complexity of DPP testing.
Paper proposes kernel-based tests for model misspecification.
problem Determining if a model is misspecified.
method Minimum distance estimators based on MMD and KSD.
result Correct test level maintained without data splitting.
New test assesses probabilistic model calibration without expensive approximations.
problem Assessing calibration of probabilistic models with scores.
method Kernel Calibration Conditional Stein Discrepancy (KCCSD) test using new score-based kernels.
result Control over type-I error with improved scalability and efficiency.
Power of network tests degrades when vertices are misaligned.
problem Power loss in network hypothesis testing due to vertex shuffling.
method Theoretical analysis and simulations of Frobenius norm differences in random dot product and stochastic block models.
result Shuffling vertices can significantly reduce the power of network tests.
The paper sets thresholds for testing correlation in hypergraphs, distinguishing between independent and correlated states.
problem Testing correlation between two hypergraphs under different models.
method Derives sharp information-theoretic thresholds for distinguishing between null and alternative hypotheses.
result The testing threshold decreases as the hypergraph's uniformity (m) increases, making correlation testing easier for higher uniformity.
We study distribution testing with communication and memory constraints in the following computational models: (1) The {\em one-pass streaming model} where the goal is to minimize the sample complexity of the protocol subject to a memory constraint, and (2) A {\em distributed model} where the data samples reside at mul…
We propose two nonparametric statistical tests of goodness of fit for conditional distributions: given a conditional probability density function p(y∣x) and a joint sample, decide whether the sample is drawn from p(y∣x)rx(x) for some density rx. Our tests, formulated with a Stein operator, can be applied to any…
Tent adapts models during testing by minimizing entropy of predictions.
problem Adapting models to new data during testing with limited information.
method Test entropy minimization (tent) and online channel-wise affine transformations.
result Reduces generalization error on various datasets and benchmarks.
A goodness-of-fit test for DCSBM improves scalability and power for large sparse networks.
problem Testing goodness-of-fit for degree-corrected stochastic block models (DCSBM) in large sparse networks.
method Proposes an adjusted chi-square test statistic for multinomial distributions, adjusted for degree-corrected networks, and applies it to compressed adjacency matrices.
result The test statistic converges in distribution under null, and is consistent in recovering the number of communities.