Paper introduces methods for more reliable probabilistic predictions with confidence intervals.
problem Inaccurate labeling of datasets due to unreliable probabilistic predictions from weak labeling functions.
method Proposes a methodology to provide confidence intervals for label probabilities using uncertainty sets of distributions.
result Improves reliability of probabilistic predictions and provides confidence intervals for label probabilities.
Paper tackles multi-source transfer learning with diverse labeling volume and reliability.
problem Challenges in multi-source transfer learning with diverse labeling volume and reliability.
method Combines domain similarity and source reliability through a new transfer learning method, and integrates distribution matching and uncertainty sampling in pool-based active learning.
result Demonstrates superior performance over state-of-the-art transfer learning methods.
Supervised method improves crowdsourcing accuracy.
problem Unsupervised crowdsourcing ignores labeler reliability.
method Saddle point algorithm to identify reliable labelers.
result Supervised method outperforms unsupervised methods.
Deep models can fit noisy labels, but robustness and reliability are still issues.
problem Training deep models with noisy labels leads to unreliable uncertainty quantification.
method Analysis of conditional distribution over noisy labels and evaluation of robust loss functions.
result Strictly proper and robust loss functions preserve accuracy but do not guarantee reliability.
PAS method improves UDA by progressively refining subspaces for reliable pseudo-labels.
problem Mitigating mode collapse in unsupervised domain adaptation.
method Progressive Adaptation of Subspaces (PAS) approach.
result PAS effectively mitigates mode collapse and improves performance in UDA and PDA.
New method improves semi-supervised learning with missing labels.
problem Reliable classification with missing labels in semi-supervised learning.
method Develops a new semi-supervised learning approach that relaxes assumptions about unlabeled data.
result Provides classifiers that reliably quantify label uncertainty.
POTA improves short text clustering by generating reliable pseudo-labels.
problem Limited discriminative representations in short texts.
method POTA uses instance-level attention and optimal transport for semantic consistency and cluster structure.
result POTA outperforms state-of-the-art methods in short text clustering.
SMART simplifies data labeling for machine learning.
problem Creating labeled training data for machine learning.
method Intuitive web interface, active learning, inter-rater reliability.
result Reduces the need for labeled data and improves label quality.
Study shows resampling labels improves classifier performance in noisy data.
problem Balancing sample size vs label reliability in noisy data.
method Comparing different validation strategies and analyzing MNIST database with varying noise levels.
result Classifier performance declines with high incorrect labels, highlighting the importance of resampling.
Improves reliability of medical diagnosis uncertainty estimates.
problem Label uncertainty in medical diagnosis.
method Post-hoc alpha-calibration method for neural network classifiers. result Significantly enhances reliability of uncertainty estimates.
Method cleans noisy training labels for biomedical data.
problem Accurately labeling biomedical data is challenging.
method Reliability-based training data cleaning with inductive conformal prediction.
result Significant enhancements in classification performance across multiple tasks.
MLDemon monitors ML systems post-deployment, improving reliability with real-time performance estimates and expert labels.
problem Ensuring reliability of machine learning systems post-deployment, especially when user inputs differ from training data.
method Integrates unlabeled and on-demand labeled data to monitor ML model performance in real-time, deciding when to acquire expert labels.
result Outperforms existing approaches in temporal datasets with diverse distribution drifts, providing theoretical optimality for distribution drifts.
Proposes efficient calibration for indoor localization models.
problem Calibration data scarcity in wireless indoor localization.
method Uses synthetic labels and prediction sets to fine-tune a predictor and estimate bias.
result Yields rigorous coverage guarantees for prediction sets.
Study efficient interactive learning for structured outputs with reliable computation.
problem Interactive learning with noisy labels and structured output spaces.
method Identify and utilize CRISPs (probabilistic models) that guarantee reliable and efficient computation of probabilistic quantities.
result CRISPs enable robust and efficient active and skeptical learning in large structured output spaces.
Triangular Estimation optimally estimates worker reliability in noisy crowdsourcing.
problem Accurately estimating the reliability of workers in crowdsourcing systems.
method Proposes Triangular Estimation (TE) for estimating worker reliability, proving minimax optimality.
result Triangular Estimation is minimax optimal and matches theoretical lower bounds.
Estimates calibration error under label shift without labels.
problem Ensuring model reliability in the face of dataset shift without access to labels.
method Importance re-weighting of the labeled source distribution to estimate calibration error under label shift.
result Effective and reliable CE estimation with respect to the shifted target distribution.
Paper tackles deep learning from crowdsourcing, improving neural network training.
problem Learning deep neural networks from noisy, unreliable crowd labels.
method Developed an EM algorithm for annotator reliability estimation and a crowd layer for end-to-end training.
result Achieved state-of-the-art results across various crowdsourced datasets.
Majority Vote is optimal for reliable data labeling under certain conditions.
problem Reliable data labeling requires aggregating multiple annotators' labels, but the optimality of Majority Vote is not well understood.
method Characterized conditions under which Majority Vote achieves the optimal label estimation error.
result Majority Vote optimally recovers labels for a given class distribution under tolerable annotation noise limits.
Fairness audits fail under missing protected labels, especially at zero access.
problem Understanding the reliability of fairness audits with incomplete protected-label data.
method Introduced a seed-calibrated stress test to separate missingness effects from seed-to-seed movement.
result Missing protected labels do not significantly alter fairness mitigation methods, but they can lead to harmful intersectional outcomes.
The paper proposes a method to reliably select design algorithms for machine learning-guided design tasks.
problem Choosing the right design algorithm for machine learning-guided design tasks.
method Combining designs' predicted property values with held-out labeled data to reliably forecast characteristics of the label distributions produced by different design algorithms.
result The method is guaranteed to return design algorithms that yield successful label distributions.
Anytime-valid confirmation of label-shift corrections
problem Small-batch scientific deployments with scarce labeled outcomes
method Conditional e-value and martingale-based rule
result Nonnegative martingale and anytime-valid confirmation rule
OpinionRank uses graph-based ranking to improve unreliable crowdsourced labels.
problem Improving trustworthiness of crowdsourced labels for machine learning.
method Graph-based spectral ranking to integrate unreliable labels.
result OpinionRank outperforms conventional algorithms in reliability and scalability.
New approach uses under-trained deep ensembles to learn from noisy labels.
problem Improper labelling hinders reliable generalization in supervised learning.
method Under-trained deep ensembles, each trained on a subset of data, combine to form better labels.
result Significant performance improvement in accuracy and kappa for noisy label tasks.
Paper introduces active Bayesian method for assessing black-box classifiers efficiently.
problem Need to assess performance of black-box classifiers reliably with limited labels.
method Develops inference strategies and proposes active Bayesian framework for efficient instance selection.
result Significant gains in performance assessment with fewer labels compared to traditional methods.
The paper studies how to allocate human validation in AI-assisted tasks to minimize errors.
problem Heterogeneous reliability of AI-generated signals across tasks, products, and customer segments.
method Tuned prediction-powered inference, upper confidence bounds policy, Neyman square-root rule.
result The proposed policy outperforms uniform and epsilon-greedy allocation, closing most of the gap to the oracle when reliability is heterogeneous.
Symmetric losses improve classifier robustness from corrupted labels.
problem Improving classifier performance from corrupted labels.
method Symmetric losses that satisfy a certain condition.
result Symmetric losses enhance robust classification from corrupted labels.
A new framework evaluates LLMs by considering judge reliability.
problem Evaluating LLMs without ground truth labels can lead to biased results.
method Introduces judge-specific discrimination parameters and estimates model quality and judge reliability.
result Improves agreement with human preferences and produces calibrated uncertainty quantification.
Study integrates reliability constraints into generation planning models.
problem Challenges in integrating reliability constraints with generation planning models.
method Leverages a weighted oblique decision tree (WODT) technique to embed reliability verification constraints.
result Demonstrates effectiveness in achieving reliable and optimal planning solutions.
BCCP uses bandit feedback to provide reliable predictions with limited labeled data.
problem Limited labeled data and bandit feedback challenge online set-valued classification.
method BCCP uses stochastic gradient descent to train model and make set-valued inferences with unbiased estimation of true label.
result BCCP offers coverage guarantees on a class-specific granularity.
Paper improves short text clustering by integrating semantic relationships into Optimal Transport.
problem Erroneous pseudo-labels caused by neglecting semantic consistency in existing OT methods.
method Designs an instance-level attention mechanism to capture semantic relationships and integrates them into the OT formulation.
result Generates reliable pseudo-labels that improve clustering accuracy.
New algorithm reliably learns ReLU functions efficiently.
problem Learning ReLU functions in the Reliable Agnostic model.
method Combines kernel methods, polynomial approximations, and dual-loss approach.
result First efficient algorithms for reliable learning of ReLU functions.
Probabilistic SR method speeds up high-fidelity simulations with reliable uncertainty estimates.
problem Lack of reliable uncertainty quantification in deep-learning based SR methods.
method Statistical Finite Element Method and energy-based generative modeling.
result Efficient high-resolution predictions with inherent uncertainty estimates.
Boosting for label ranking outperforms existing methods.
problem Improving label ranking predictions using boosting techniques.
method Proposed a boosting algorithm tailored for label ranking tasks.
result Significantly outperforms existing label ranking algorithms.
ease.ml/ci integrates machine learning models rigorously with minimal labeling effort.
problem Lack of rigorous continuous integration for machine learning models.
method Design of a domain-specific language and novel optimizations to reduce labeling effort.
result Achieved single accuracy point error tolerance with 0.999 reliability using minimal labeling effort.
Paper proposes unsupervised learning for optimizing deep neural networks in time-sensitive applications.
problem Inaccurate optimization solutions from supervised learning for real-time applications.
method Proposes unsupervised deep learning to learn latent function with optimization constraints as supervision.
result DNN can optimize variable problems in ultra-reliable communications without supervision labels.
New method reduces labeler costs by aggregating predictions from local classifiers.
problem Reduce labeler costs in multiclass classification.
method Model K-class classification using smaller classifiers trained on subsets of tasks. result Near-optimal scheme for designing classifier configurations reduces labeler costs.
Self-training improves generalization by fitting to reliable pseudo-labels or gradually improving the classification plane.
problem Understanding how self-training improves generalization in high-dimensional Gaussian mixtures.
method Analyzing iterative self-training on binary Gaussian mixtures in the asymptotic limit.
result ST improves generalization by fitting to reliable pseudo-labels or gradually improving the classification plane.
DUPLE tackles cross-deployment recognition in fiber-optic perimeter security with meta-learning.
problem Cross-deployment recognition challenges in fiber-optic perimeter security due to label scarcity and distribution shifts.
method DUPLE employs statistically guided meta-learning to enhance recognition robustness across unseen deployments.
result DUPLE consistently outperforms traditional and meta-learning baselines in cross-deployment DFOS benchmarks.
New algorithms learn from comparisons to classify data robustly to noise.
problem Learning robust classifiers from noisy data efficiently.
method Introducing comparison queries to active learning, providing noise-tolerant classifiers.
result First time and query efficient algorithms for robust learning under bounded noise.
New method improves reliability of selecting individuals based on predicted treatment effects.
problem Reliability of selecting individuals based on predicted conditional average treatment effects (CATE) is unreliable.
method Denoised Conformal Alignment, combining proxy errors, variance estimation, and Benjamini-Hochberg selection.
result Significantly improved power in selecting individuals while maintaining false discovery rate control.
New methods identify concepts in trained embeddings reliably without human labels.
problem Identifying interpretable concepts in trained embedding spaces without human labels.
method Explicitly connecting concept discovery to PCA and ICA, proposing novel approaches for dependent concepts.
result Proven methods outperform competitors on a variety of experiments, achieving up to 29% better alignment with ground truth.
Efficiently classifies binary labels with XOR queries, even under noisy conditions.
problem Binary classification with unknown labels using XOR queries.
method Effective query type and an efficient inference algorithm for noisy conditions.
result Achieves information-theoretic limit on optimal number of queries.
Limited supervision can enable reliable disentangled representation learning.
problem Learning disentangled representations without inductive biases is theoretically impossible.
method Investigated the impact of limited supervision (0.01--0.5% of data) on disentanglement methods.
result A small number of labeled examples (0.01--0.5\% of the data set) is sufficient for model selection.
IUPM monitors machine learning models under gradual shifts using optimal transport and active labeling.
problem Gradual distribution shifts lead to unnoticed accuracy declines in machine learning models.
method Incremental Uncertainty-aware Performance Monitoring (IUPM) using optimal transport and active labeling.
result IUPM outperforms existing baselines in gradual shift scenarios and guides label acquisition more effectively.
Unified framework for certifying LLM reliability without extra supervision.
problem Improving reliability of large language models without additional supervision.
method Unified framework using majority voting and Martingale Majority Certificate (MMC).
result Certifiable inference in LLMs with statistical guarantees and adaptive stopping rules.
The paper introduces a method for multi-label classification that allows partial predictions.
problem Handling multi-label classification with the option to abstain from predictions.
method Formalized MLC with abstention as a generalized loss minimization problem.
result Initial results for Hamming loss, rank loss, and F-measure.
Active learning performance hard to assess during deployment.
problem Difficulty in estimating active learning performance during deployment.
method Formalized task, identified challenges, evaluated existing strategies.
result No reliable method to estimate active learning performance during training.
Ask-n-Learn uses gradient embeddings for active learning in image classification.
problem Efficiently labeling large amounts of training data for deep models.
method Gradient embeddings based on pseudo-labels, prediction calibration, and data augmentation.
result Significant improvements over state-of-the-art baselines on image classification tasks.