DivDis learns diverse hypotheses from underspecified data to improve robustness.
problem Learning from underspecified datasets leads to multiple equally viable solutions, causing out-of-distribution issues.
method DivDis framework: 1) learns diverse hypotheses using unlabeled test data, 2) selects one hypothesis with minimal additional supervision.
result DivDis finds robust features in image and natural language processing problems.
We consider the problem of learning from sparse and underspecified rewards, where an agent receives a complex input, such as a natural language instruction, and needs to generate a complex response, such as an action sequence, while only receiving binary success-failure feedback. Such success-failure rewards are often …
Underspecified ML models can behave unpredictably in real-world use.
problem ML models can fail in real-world deployment due to ambiguous predictors.
method Identified underspecification as the cause, showing it affects various ML domains.
result Underspecified models can behave differently in deployment domains.
New method uses few instruments to estimate complex causal effects.
problem Estimating causal effects with limited instruments in high-dimensional settings.
method Sequentially selects and combines instruments to estimate the treatment effect.
result Can reliably recover the treatment effect's projection onto the instrumented subspace.
Different optimizer choices lead to different financial model predictions.
problem The impact of optimizer choice on neural network models in financial time series.
method Analysis of large-scale volatility forecasting for S&P 500 stocks using various model-training-pipeline pairs.
result Optimizer choice reshapes non-linear response profiles and temporal dependence in financial models, leading to different functional outcomes.
We present local ensembles, a method for detecting underspecification -- when many possible predictors are consistent with the training data and model class -- at test time in a pre-trained model. Our method uses local second-order information to approximate the variance of predictions across an ensemble of models from…
Second-order optimizers retain residual information after data deletion, affecting machine unlearning.
problem Residual information in second-order optimizers after data deletion.
method Comparison of first-order and second-order learners, eigendecomposition analysis.
result Second-order optimizers retain residual information, not detectable by first-order analysis.
Clustering is an underspecified task: there are no universal criteria for what makes a good clustering. This is especially true for relational data, where similarity can be based on the features of individuals, the relationships between them, or a mix of both. Existing methods for relational clustering have strong and …
New method resolves time order in genetic mutation models.
problem Underspecification in modeling genetic mutation time evolution.
method Continuous-time Markov chains with additional independent items.
result Additional items help determine time order and resolve underspecification.
Neural networks struggle with extrapolation, but a new framework allows them to learn counterfactual invariances.
problem Neural networks' inability to extrapolate beyond training data distribution.
method Introduces a learning framework that allows neural networks to extrapolate over group transformations based on counterfactual invariances.
result Neural networks can learn counterfactual invariances from a single environment, overcoming their limitations in extrapolation.
The paper identifies causal effects in latent variable models using higher-order cumulants.
problem Challenges in identifying causal effects in latent variable models with latent confounders.
method Using higher-order cumulants, the paper addresses two challenging setups: a single proxy variable and underspecified instrumental variables.
result Causal effects are identifiable with a single proxy or instrument.
The key distinguishing property of a Bayesian approach is marginalization instead of optimization, not the prior, or Bayes rule. Bayesian inference is especially compelling for deep neural networks. (1) Neural networks are typically underspecified by the data, and can represent many different but high performing models…
Supervised machine learning models boast remarkable predictive capabilities. But can you trust your model? Will it work in deployment? What else can it tell you about the world? We want models to be not only good, but interpretable. And yet the task of interpretation appears underspecified. Papers provide diverse and s…
Bayesian deep learning improves neural network accuracy and generalization.
problem Improving accuracy and calibration of deep neural networks.
method Bayesian marginalization and deep ensembles to approximate marginalization, and tempering for calibrating predictive distributions.
result Bayesian approaches improve deep neural networks' accuracy and generalization.
New method uses limited labeled data and multiple starts to adapt models across domains.
problem Accurate predictions in target domain with few labeled data.
method Fine-tuning from multiple adaptive starts, extending UDA methods.
result Minimax-optimal target performance with limited labeled target data.
Bayesian DL model improves DCMs for better predictive and inferential performance.
problem Limited interpretability and predictive underperformance of traditional DCMs.
method Integrates deep learning with approximate Bayesian inference (SGLD).
result Improves predictive and inferential metrics in discrete choice models.
The paper proposes a method to align AI models using conformal risk control.
problem Aligning AI models to meet end-user requirements in non-generative settings.
method Post-processing a pre-trained model to better align with a subset of functions using conformal risk control.
result A probabilistic guarantee that the resulting conformal interval around a model contains a function approximately satisfying a desired property.
Most work in machine reading focuses on question answering problems where the answer is directly expressed in the text to read. However, many real-world question answering problems require the reading of text not because it contains the literal answer, but because it contains a recipe to derive an answer together with …
Proposes a new approach for domain adaptation using latent representations.
problem Handling distribution shifts between source and target domains in high-dimensional data.
method Learn compact latent representations based on the label's Markov blanket, partitioning into parents, children, and spouses.
result General domain adaptation can be achieved by learning representations of the label's parents, children, and spouses.
This paper introduces CENIE to quantify environment novelty for better UED.
problem Challenges in measuring environment novelty for effective UED.
method CENIE framework using state-action space coverage and Gaussian Mixture Models.
result CENIE improves UED performance across multiple benchmarks.
New study on time series anomaly detection shows overlapping inference improves performance.
problem Heterogeneous evaluation practices and inference procedures in time series anomaly detection.
method Unified training, tuning, and evaluation protocol on TSB-AD benchmark, analyzing overlapping vs. disjoint inference.
result Overlapping inference yields consistent improvements, with average relative gain up to +28%.
Big data sets must be carefully partitioned into statistically similar data subsets that can be used as representative samples for big data analysis tasks. In this paper, we propose the random sample partition (RSP) data model to represent a big data set as a set of non-overlapping data subsets, called RSP data blocks,…
Prevents sensitive data generation in diffusion models using labeled and unlabeled data.
problem Generating sensitive data in diffusion models using unlabeled data.
method Positive-Unlabeled Diffusion Models, approximating ELBO with labeled and unlabeled data.
result Prevents the generation of sensitive data without compromising image quality.
Study reveals Data Shapley's inconsistent performance in data selection tasks.
problem Inconsistency of Data Shapley's performance in data selection across different settings.
method Hypothesis testing framework and identification of utility functions.
result Data Shapley's performance is no better than random selection without specific constraints.
PRRO generates synthetic tabular data that improves SL performance and class distribution.
problem Low SL utility of synthetic data due to class imbalance and overlooked data relationships.
method Data pruning and column reordering to optimize SL utility.
result Synthetic data generated with PRRO enhances predictive performance and class distribution.
Defines data science as a natural ecosystem with challenges and missions.
problem Challenges and missions in data science due to 5D complexities and data life cycle phases.
method Systemic and data-centric view of data science as a fusion of data universe and its challenges, formalizing a general-purpose architecture.
result Essential data science as a natural ecosystem integrating specific disciplines and high-impact applications.
Differences in data size per class, also known as imbalanced data distribution, have become a common problem affecting data quality. Big Data scenarios pose a new challenge to traditional imbalanced classification algorithms, since they are not prepared to work with such amount of data. Split data strategies and lack o…
Synthetic data enhances analytics but requires careful volume management.
problem Accuracy of statistical methods on synthetic data vs. raw data.
method Synthetic Data Generation for Analytics framework using tabular diffusion models.
result Error rate decreases with more synthetic data but may stabilize or increase.
New test ensures quality of shared data in machine learning.
problem Ensuring quality of external data in machine learning tasks.
method Distribution-free two-sample testing procedures grounded in conformal outlier detection.
result Identifies valuable external data agents for model personalization.
Differential privacy allows quantifying privacy loss resulting from accessing sensitive personal data. Repeated accesses to underlying data incur increasing loss. Releasing data as privacy-preserving synthetic data would avoid this limitation, but would leave open the problem of designing what kind of synthetic data. W…
Paper creates fair synthetic data ensuring equal predictions across sensitive attributes.
problem Ensuring fair predictions across sensitive attributes in synthetic data.
method Equalizing target probability distributions across sensitive attributes in synthetic data generation.
result Synthetic data provides strong fair predictions, equal across all thresholds.
DPA preserves data distribution in reduced dimensions.
problem Loss of data distribution in dimension reduction.
method DPA combines encoder and decoder to match data distribution.
result DPA successfully reconstructs data distribution.
Efficient synthetic data generation improves model performance on tabular data.
problem Improving model robustness and performance with scarce or low-quality data.
method Hardness characterization to identify high-value training points, generating synthetic data only from these points.
result Synthetic data generated from hardest points outperforms non-targeted methods on tabular datasets.
For most problems in science and engineering we can obtain data sets that describe the observed system from various perspectives and record the behavior of its individual components. Heterogeneous data sets can be collectively mined by data fusion. Fusion can focus on a specific target relation and exploit directly ass…
DAERNN models censored data using neural networks with data augmentation.
problem Handling censored data in expectile regression.
method Data augmentation based Expectile Regression Neural Networks (ERNNs).
result DAERNN outperforms existing censored ERNNs methods and achieves comparable predictive performance to fully observed data.
Data preprocessing techniques are devoted to correct or alleviate errors in data. Discretization and feature selection are two of the most extended data preprocessing techniques. Although we can find many proposals for static Big Data preprocessing, there is little research devoted to the continuous Big Data problem. A…
Data stream classification methods demonstrate promising performance on a single data stream by exploring the cohesion in the data stream. However, multiple data streams that involve several correlated data streams are common in many practical scenarios, which can be viewed as multi-task data streams. Instead of handli…
Data collection is a major bottleneck in machine learning and an active research topic in multiple communities. There are largely two reasons data collection has recently become a critical issue. First, as machine learning is becoming more widely-used, we are seeing new applications that do not necessarily have enough …
This paper quantifies uncertainty in Data Shapley using statistical inference.
problem Uncertainty in data valuation due to dynamic data distribution.
method Established relationship with U-statistics and quantified uncertainty using statistical inference.
result Confidence intervals for Data Shapley estimations are provided.
DCoM uses deep neural networks to detect semantic data types from raw column values.
problem Detecting semantic data types from dirty and unseen data.
method DCoM employs multi-input NLP-based deep neural networks trained on 686,765 data columns.
result DCoM outperforms existing methods significantly on 78 different semantic data types.
Causal inference from observational data is the goal of many data analyses in the health and social sciences. However, academic statistics has often frowned upon data analyses with a causal objective. The introduction of the term "data science" provides a historic opportunity to redefine data analysis in such a way tha…
Task-agnostic data valuation without validation requirements.
problem Valuing data without specific task assumptions.
method Estimating data diversity and relevance through queries without raw data.
result Estimates capture the diversity and relevance of seller's data for the buyer.
Data mining is about obtaining new knowledge from existing datasets. However, the data in the existing datasets can be scattered, noisy, and even incomplete. Although lots of effort is spent on developing or fine-tuning data mining models to make them more robust to the noise of the input data, their qualities still st…
New algorithm improves data imputation for complex multimodal data sets.
problem Artifacts in imputation methods for multimodal distributions.
method Combines kNN and KDE for probabilistic estimates. result Lower imputation errors and higher likelihood estimates.
New algorithms for clustering and synthetic data generation of heterogeneous tabular datasets.
problem Clustering and generating synthetic data from heterogeneous tabular datasets with hidden cluster structure.
method Developed MMM and MMMsynth algorithms for clustering and synthetic data generation.
result MMMsynth algorithm outperforms other literature tabular-data generators and approaches real data performance.
WeMix improves data augmentation by correcting bias in deep learning.
problem Data augmentation's effectiveness is limited by data bias.
method Developed AugDrop and MixLoss algorithms to correct data bias.
result WeMix improves data augmentation performance through bias correction.
A new method reduces data valuation variance for more trustworthy data trading.
problem Data valuation and trustworthy data trading in algorithmic prediction.
method Variance reduced Shapley value estimation using stratified sampling.
result VRDS method reduces estimation variance and improves data marketplace development.
Adapts data analysis for growing data, improving generalization guarantees.
problem Challenges of overfitting and statistical validity in adaptive workflows with growing data.
method Generalizes adaptive analysis on dynamic data, incorporating time-varying empirical accuracy bounds and mechanisms.
result First generalization bounds for adaptive analysis on dynamic data, matching prior works' improvement over data splitting.