Study evaluates methods for improving model robustness to various real-world distribution shifts.
problem Improving model robustness to real-world distribution shifts like geographic changes.
method Introduced new datasets and evaluated existing methods on four types of shifts (style, blurriness, location, camera operation).
result Data augmentations and larger models can improve robustness on real-world distribution shifts, contrary to prior claims.
GANs can learn hierarchical distributions in real-world images efficiently.
problem Understanding and efficiently learning complex, real-world distributions with GANs.
method Formally studying how GANs can learn hierarchically generated distributions close to real-life image distributions using SGDA.
result Training GANs via SGDA can efficiently learn distributions with a 'forward super-resolution' structure, both in sample and time complexities.
Study evaluates large language models' ability to understand probabilistic real-world distributions.
problem Understanding how LLMs grasp probabilistic knowledge of real-world distributions.
method Developed a benchmark to test LLMs' ability to learn and represent empirical distributions across various domains.
result LLMs perform poorly in understanding real-world statistics and do not naturally internalize these distributions.
Reliable uncertainty estimates are an important tool for helping autonomous agents or human decision makers understand and leverage predictive models. However, existing approaches to estimating uncertainty largely ignore the possibility of covariate shift--i.e., where the real-world data distribution may differ from th…
New algorithms improve federated learning accuracy and stability with real-world data.
problem Real-world data diversity and imbalance challenge federated learning.
method Developed new algorithms (FedVC, FedIR) to resample and reweight data.
result Significant improvements in accuracy and stability of federated learning.
Study evaluates scalability and real-world impact of disentangled representations.
problem Scalability and real-world impact of disentangled representations.
method New high-resolution dataset and architectures for disentangled representation learning.
result Disentanglement predicts out-of-distribution task performance.
Robust CD method for real-world time series with power-law distributions.
problem Challenges in causal discovery due to noise sensitivity.
method Power-law spectral feature extraction for robust CD.
result Consistently outperforms state-of-the-art alternatives on real-world datasets.
New method estimates treatment effects across different populations.
problem Estimating treatment effects across populations with changing distributions.
method SBRL-HAP framework combining balancing and independence regularizers with hierarchical attention.
result Significant improvement in HTE estimation across out-of-distribution populations.
Improves reinforcement learning policies for robustness.
problem Lack of robustness in reinforcement learning policies.
method Risk-aware Distributional Reinforcement Learning (SDPG) with CVaR.
result Risk-averse policies achieve robustness against disturbances.
We created financial benchmarks for distribution shifts in crude oil prices and volatility.
problem Scarcity of task-labeled time-series benchmarks in finance.
method Transformed asset price data into volatility proxies, generated task labels based on distribution shifts, and made datasets publicly available.
result Inclusion of task labels improves continual learning algorithms' performance on real-world data.
Mitigates anomaly score imbalance in long-tailed distributions.
problem Class imbalance in normal data leads to skewed anomaly detection performance.
method Proposes an importance-weighted loss function to balance anomaly scores.
result Improves anomaly detection performance by 0.043 on real-world datasets.
Method learns conditional distributions using neural entropic optimal transport.
problem Challenges in learning multiple conditional distributions.
method Neural entropic optimal transport method with two networks and regularization.
result Effective learning of conditional distributions with limited samples.
Recently, reinforcement learning (RL) algorithms have demonstrated remarkable success in learning complicated behaviors from minimally processed input. However, most of this success is limited to simulation. While there are promising successes in applying RL algorithms directly on real systems, their performance on mor…
TAET tackles long-tailed distributions in adversarial robustness.
problem Long-tailed distributions complicate adversarial robustness in real-world applications.
method TAET integrates an initial stabilization phase followed by a stratified equalization adversarial training phase.
result TAET achieves significant improvements in robustness and efficiency.
GRAM enhances deep RL for reliable real-world deployment.
problem Generalizing deep RL across in-distribution and out-of-distribution scenarios.
method Introduces a robust adaptation module and a joint training pipeline.
result GRAM achieves strong generalization performance in simulations and hardware.
Paper tackles online adaptation to changing label distributions.
problem Adapting machine learning models to changing label distributions in real-world settings.
method Leverages novel analysis to show estimation of expected test loss is possible without true labels. Proposes adaptation algorithms inspired by classical online learning techniques.
result Empirically verified that OGD is particularly effective and robust to various label shift scenarios.
New RESK distributions improve robust clustering of skewed data.
problem Robustly clustering non-symmetric, heavy-tailed data clusters.
method Proposes RESK distributions and an EM algorithm with robust skew-Huber M-estimator.
result Numerical experiments confirm the effectiveness of the proposed methods.
Research on deep learning generalization in real-world applications.
problem Understanding and improving deep learning generalization in non-i.i.d. real-world data.
method Analyzing deep net generalization, identifying and addressing assumptions and problem settings failures.
result Proposes methods to address failures in deep net generalization for real-world applications.
This paper improves change-point detection for complex data streams using denoising score matching.
problem Timely identification of distributional shifts in high-dimensional, complex data streams.
method Score-based CUSUM change-point detection with denoising score matching.
result Denoising score matching enhances detection power by effectively controlling noise scale.
Bayesian optimization adapts domain parameters for more robust robot policies.
problem Learning policies for robot control from simulation data often fails in the real world due to the 'reality gap'.
method Bayesian Domain Randomization (BayRn) uses Bayesian optimization to adapt domain parameter distributions during training.
result BayRn achieves better sim-to-real transfer compared to fixed distribution methods.
The study diagnoses fairness issues in healthcare models under distribution shifts.
problem Understanding and diagnosing fairness changes in machine learning models under distribution shifts in healthcare.
method Causal framing and conditional independence tests to characterize distribution shifts.
result Knowledge of distribution shifts helps diagnose fairness transfer failures, including complex cases.
The Poisson distribution has been widely studied and used for modeling univariate count-valued data. Multivariate generalizations of the Poisson distribution that permit dependencies, however, have been far less popular. Yet, real-world high-dimensional count-valued data found in word counts, genomics, and crime statis…
Improved measure of predictive uncertainty for machine learning models.
problem Current measure of predictive uncertainty assumes BMA predictive distribution is equivalent to true model's distribution.
method Introduced a new measure based on information theory to correct the assumption.
result Our measure behaves more reasonably in synthetic tasks and is advantageous in real-world applications.
Study evaluates conformal prediction methods for safety in vision models under shifts and long-tailed data.
problem Safety guarantees of conformal prediction methods under distribution shifts and long-tailed data.
method Empirical evaluation of post-hoc and training-based conformal prediction methods on large-scale datasets and models.
result Performance of conformal prediction methods degrades significantly under distribution shifts and long-tailed data.
Multi-instance learning (MIL) deals with tasks where data is represented by a set of bags and each bag is described by a set of instances. Unlike standard supervised learning, only the bag labels are observed whereas the label for each instance is not available to the learner. Previous MIL studies typically follow the …
Nowadays, machine learning methods have been widely used in stock prediction. Traditional approaches assume an identical data distribution, under which a learned model on the training data is fixed and applied directly in the test data. Although such assumption has made traditional machine learning techniques succeed i…
AdapTable adapts tabular models to shifts without source data, improving HELOC performance.
problem Distribution shifts in tabular data threaten model performance.
method Shift-aware uncertainty calibrator and label distribution handler.
result Up to 16% improvement on HELOC dataset.
Method converts age labels into distributions to improve speaker age estimation.
problem Label ambiguity in age labels makes precise speaker age estimation challenging.
method Converts age labels into label distributions and uses label distribution learning.
result Our method outperforms baseline methods by reducing MAE by 10% on a real-world dataset.
Study reveals how spectral bias affects learnability on real-world data.
problem Understanding how well complex datasets can be learned using kernel methods.
method Use eigenvalues and eigenfunctions from idealized data to reveal spectral bias on real-world data.
result Bound learnability on real-world data using symmetries of realistic kernels.
A new method for incorporating preferences in multi-objective Bayesian optimization.
problem Incorporating preferences in computationally expensive multi-objective optimization problems.
method Building independent surrogate models on each objective function and using Generalised value distribution to approximate the scalarizing function.
result The proposed multi-surrogate approach outperforms the mono-surrogate approach on benchmark and real-world problems.
PH-VAE models heavy-tailed data with flexible Phase-Type distributions.
problem Standard VAEs fail to capture heavy-tailed behavior in real-world data.
method PH-VAE uses Phase-Type distributions defined by continuous-time Markov chains to adaptively model tail behavior.
result PH-VAE significantly outperforms existing heavy-tail-aware VAEs in approximating diverse heavy-tailed distributions.
Bayesian optimization tackles uncertainty in context variables.
problem Sequential decision-making under context distributional uncertainty.
method Wasserstein Distributionally Robust Bayesian Optimization.
result Sublinear regret bounds matching state-of-the-art results.
Proposes a new method for nonlinear models with robustness guarantees.
problem Distributional robustness in nonlinear models with causality.
method Representation learning and identifiable representation learning.
result First causality-inspired robustness method with finite-radius guarantees in nonlinear settings.
New research finds uncertainty estimation techniques fail to reliably detect abnormal medical cases.
problem Uncertainty estimation does not reliably detect out-of-distribution patients in medical tabular data.
method A series of tests on various uncertainty estimation techniques on real-world medical data.
result Almost all techniques fail to identify out-of-distribution patients, contradicting earlier findings.
Study quantifies distribution shifts and uncertainties to improve machine learning model robustness.
problem Distribution shifts between training and test datasets impact model generalization and robustness.
method Synthetic data generation and quantitative measures (KL divergence, JS distance, Mahalanobis distance) to assess data similarity and model uncertainty.
result Utilizing statistical measures like Mahalanobis distance helps assess distribution shift and model uncertainty.
New model tackles real-world distribution mismatches in machine learning.
problem Real-world applications often have training and test distributions that differ.
method Developed a learning model based on information theory using importance sampling.
result The model performs better under large distribution deviations.
Distributed Quantum Gaussian Processes improve modeling in multi-agent systems.
problem Limited expressivity of classical kernels in complex domains.
method Distributed Quantum Gaussian Process (DQGP) with DR-ADMM algorithm.
result Enhanced modeling capabilities and scalability in multi-agent systems.
DDG-DA predicts future data distribution to adapt models for predictable concept drift.
problem Adapting models to streaming data with predictable concept drift.
method Train a predictor to forecast future data distribution, generate training samples, and train models on them.
result Significant improvement on multiple models in real-world tasks.
In retailer management, the Newsvendor problem has widely attracted attention as one of basic inventory models. In the traditional approach to solving this problem, it relies on the probability distribution of the demand. In theory, if the probability distribution is known, the problem can be considered as fully solved…
Paper addresses challenges in benchmarking stream learning algorithms with real-world data.
problem Lack of publicly available non-stationary real-world datasets for evaluating stream algorithms.
method Proposes a new public data repository for benchmarking stream algorithms with real-world data.
result Mitigates problems related to dataset choice in experimental evaluation of stream classifiers and drift detectors.
Recurrent neural networks (RNNs) have been extraordinarily successful for prediction with sequential data. To tackle highly variable and noisy real-world data, we introduce Particle Filter Recurrent Neural Networks (PF-RNNs), a new RNN family that explicitly models uncertainty in its internal structure: while an RNN re…
We investigate the non-identifiability issues associated with bidirectional adversarial training for joint distribution matching. Within a framework of conditional entropy, we propose both adversarial and non-adversarial approaches to learn desirable matched joint distributions for unsupervised and supervised tasks. We…
Proposes ITISC for clustering with minimized worst-case expected distortions.
problem Real-world clustering data distribution mismatch.
method Information theoretical importance sampling, constrained minimax optimization, Lagrange method.
result Validation of ITISC on synthetic and real-world datasets.
Method provides formal guarantees for decomposing model uncertainty.
problem Decomposing model uncertainty into aleatoric and epistemic components.
method Higher-order calibration using k-snapshots.
result Formal guarantees for aleatoric uncertainty matching real-world distribution.
Optimizes control interventions in real-world networks using deep-learning and network science.
problem Optimizing control over socioeconomic networks subject to constraints.
method Integrates optimization tools from deep-learning with network science.
result Characterizes vulnerability of corporate networks to takeovers.
Frengression models causal data flexibly and faithfully.
problem Challenges in robust benchmarking and evaluation of causal inference with real-world data.
method Introduces frengression, a deep generative model for joint distribution of covariates, treatments, and outcomes.
result Frengression provides accurate estimation and flexible simulation of multivariate, time-varying data.
(The third edition corrects minor typos and adds 3 chapters synthesized from published papers plus an appendix on maximum entropy distributions.) The monograph investigates the misapplication of conventional statistical techniques to fat tailed distributions and looks for remedies, when possible. Switching from thin ta…
Domain randomization (DR) is a successful technique for learning robust policies for robot systems, when the dynamics of the target robot system are unknown. The success of policies trained with domain randomization however, is highly dependent on the correct selection of the randomization distribution. The majority of…