arXiv research

A new framework for paired-sample testing in high-dimensional data.

New sampling methods improve Shapley values for explaining machine learning predictions.

Obtaining common representations from different modalities is important in that they are interchangeable with each other in a classification problem. For example, we can train a classifier on image features in the common representations and apply it to the testing of the text features in the representations. Existing m…

A new estimator, OddSHAP, simplifies Shapley value computation by focusing on odd components.

Estimating mutual information is an important statistics and machine learning problem. To estimate the mutual information from data, a common practice is preparing a set of paired samples $\{(\mathbf{x}_i,\mathbf{y}_i)\}_{i=1}^n \stackrel{\mathrm{i.i.d.}}{\sim} p(\mathbf{x},\mathbf{y})$. However, in many situations, it…

Domain adaptation aims to assist the modeling tasks of the target domain with knowledge of the source domain. The two domains often lie in different feature spaces due to diverse data collection methods, which leads to the more challenging task of heterogeneous domain adaptation (HDA). A core issue of HDA is how to pre…

We address the problem of image translation between domains or modalities for which no direct paired data is available (i.e. zero-pair translation). We propose mix and match networks, based on multiple encoders and decoders aligned in such a way that other encoder-decoder pairs can be composed at test time to perform u…

L-Cool improves image and language translation by cooling low-density samples.

Paper presents a method to align unpaired samples across different modalities.

We extend multi-way, multivariate ANOVA-type analysis to cases where one covariate is the view, with features of each view coming from different, high-dimensional domains. The different views are assumed to be connected by having paired samples; this is a common setup in recent bioinformatics experiments, of which we a…

Data augmentation is a widely used technique in many machine learning tasks, such as image classification, to virtually enlarge the training dataset size and avoid overfitting. Traditional data augmentation techniques for image classification tasks create new samples from the original training data by, for example, fli…

Likelihood-free methods perform parameter inference in stochastic simulator models where evaluating the likelihood is intractable but sampling synthetic data is possible. One class of methods for this likelihood-free problem uses a classifier to distinguish between pairs of parameter-observation samples generated using…

Weak supervision enables learning causal representations from unstructured data.

A Triangle Generative Adversarial Network ($Δ$-GAN) is developed for semi-supervised cross-domain joint distribution matching, where the training data consists of samples from each domain, and supervision of domain correspondence is provided by only a few paired samples. $Δ$-GAN consists of four neural networks, two ge…

Traditional online map tiles, widely used on the Internet such as Google Map and Baidu Map, are rendered from vector data. Timely updating online map tiles from vector data, of which the generating is time-consuming, is a difficult mission. It is a shortcut to generate map tiles in time from remote sensing images, whic…

Paper tackles missing data in irregularly-sampled time series.

A model for hit song prediction can be used in the pop music industry to identify emerging trends and potential artists or songs before they are marketed to the public. While most previous work formulates hit song prediction as a regression or classification problem, we present in this paper a convolutional neural netw…

This research shows how to learn shared representations from unpaired data.

Filtering data with a pre-trained model improves multimodal contrastive learning performance.

LSDM uses unpaired data to match latent space distributions for generative modeling.

SPLICE method disentangles shared and private latent variables from multi-view data.

The spatio-temporal graph learning is becoming an increasingly important object of graph study. Many application domains involve highly dynamic graphs where temporal information is crucial, e.g. traffic networks and financial transaction graphs. Despite the constant progress made on learning structured data, there is s…

Develops a framework for multi-objective learning in diffusion models with limited labeled data.

This paper speeds up uncertainty quantification in inverse problems using conditional normalizing flows.

Many methods for automated software test generation, including some that explicitly use machine learning (and some that use ML more broadly conceived) derive new tests from existing tests (often referred to as seeds). Often, the seed tests from which new tests are derived are manually constructed, or at least simpler t…

A family of maximum mean discrepancy (MMD) kernel two-sample tests is introduced. Members of the test family are called Block-tests or B-tests, since the test statistic is an average over MMDs computed on subsets of the samples. The choice of block size allows control over the tradeoff between test power and computatio…

USP test improves on Pearson's chi-squared and $G$-test for independence.

E-C2ST uses E-values for high-dimensional data two-sample tests.

This paper provides a comprehensive survey of Machine Learning Testing (ML testing) research. It covers 144 papers on testing properties (e.g., correctness, robustness, and fairness), testing components (e.g., the data, learning program, and framework), testing workflow (e.g., test generation and test evaluation), and …

Cheap permutation tests speed up distribution testing without sacrificing accuracy.

DRIFT uses RL to automate functional software testing efficiently.

Optimizes two-sample tests for non-Euclidean domains using spectral regularization.

Robust test for distributions under Hellinger distance, simpler than optimal tests.

A new method for kernel tests without data splitting increases power.

Framework for online hypothesis testing across various data types.

Model-X test detects conditional independence in streaming data.

Simple methods combine statistical tests for out-of-distribution detection.

Unified score and distance-based GoF tests for model adequacy.

AutoML simplifies two-sample tests for detecting distribution shifts.

New methods for CI testing under model misspecification.

New KCM tests improve specification testing via RKHS.

Discusses MultiFIT for multivariate dependence, comparing it to HSIC tests.

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 …

Develops hypothesis tests for conditional distributions using learning-theoretic bounds.

Develops tests for conditional symmetry under group actions.

Paper proposes a new method for more accurate group testing of infected patients.

This paper introduces differentially private permutation tests for hypothesis testing.

Improved MMD test for two-sample testing with random Fourier features.