RoPAWS improves semi-supervised learning on uncurated data.
problem Efficiency of semi-supervised learning with real-world unlabeled data.
method Reinterprets PAWS as a generative classifier and calibrates predictions using densities of labeled and unlabeled data.
result Significant improvement in performance for uncurated data.
WILDS 2.0 expands benchmark datasets for unsupervised adaptation.
problem Leveraging unlabeled data for distribution shifts in real-world applications.
method Curated unlabeled data across various applications, tasks, and modalities.
result State-of-the-art methods perform poorly on WILDS datasets.
While deep learning has been incredibly successful in modeling tasks with large, carefully curated labeled datasets, its application to problems with limited labeled data remains a challenge. The aim of the present work is to improve the label efficiency of large neural networks operating on audio data through a combin…
Learning algorithms that aggregate predictions from an ensemble of diverse base classifiers consistently outperform individual methods. Many of these strategies have been developed in a supervised setting, where the accuracy of each base classifier can be empirically measured and this information is incorporated in the…
We explain SSL objectives as log-likelihoods in a data curation model.
problem Lack of understanding of SSL objectives as log-likelihoods.
method Formulate SSL objectives as a log-likelihood in a generative model of data curation.
result SSL methods can be understood as lower-bounds on a principled log-likelihood.
The paper resolves the paradox of using less data in machine learning.
problem The paradox of using less data in machine learning.
method Theoretical framework and data curation strategies.
result Small curated datasets can outperform full datasets under certain conditions.
The paper shows how curated synthetic data can optimize human preferences in generative models.
problem Contamination of web-scale datasets by synthetic data affects future model training.
method Theoretical study of iterated retraining of generative models with curated synthetic data.
result Data curation can be seen as an implicit preference optimization mechanism, maximizing expected reward.
Curating labeled training data has become the primary bottleneck in machine learning. Recent frameworks address this bottleneck with generative models to synthesize labels at scale from weak supervision sources. The generative model's dependency structure directly affects the quality of the estimated labels, but select…
DataRater learns which data points are most valuable for training models.
problem Training model efficiency depends on high-quality training data.
method Meta-learning to estimate the value of data points for training.
result Meta-learning improves compute efficiency by filtering data effectively.
The paper compares aggregated data labels in curated and random bags for machine learning models.
problem Protecting user privacy in machine learning systems with aggregated data.
method Examined curated and random bags for training machine learning models and compared their performance.
result Gradient-based learning can be performed on aggregated data without performance degradation.
The paper examines how decentralized credit curators have taken over risk management from traditional protocols.
problem Risk management in decentralized credit has shifted from centralized protocols to decentralized curators.
method Analysis of ERC 4626 vaults and third-party curators, focusing on capital utilization, concentration, and fee margins.
result Curators have a significant impact on the risk profile of decentralized credit systems, with a small set of curators handling a disproportionate share of system TVL.
This research creates and classifies datasets for Setswana and Sepedi news headlines.
problem Challenges in creating and preparing datasets for low-resourced languages.
method Investigates an approach for data augmentation tailored to low resource languages.
result Improves classification performance on news topic classification task.
ConCuR generates high-quality CUDA kernels with concise reasoning traces.
problem Scarce high-quality data for kernel generation.
method Developed a pipeline to generate and curate high-quality CUDA kernels with reasoning traces.
result Our model achieves significant improvements in KernelBench setup.
Machine learning research depends on objectively interpretable, comparable, and reproducible algorithm benchmarks. We advocate the use of curated, comprehensive suites of machine learning tasks to standardize the setup, execution, and reporting of benchmarks. We enable this through software tools that help to create an…
For a social networking service to acquire and retain users, it must find ways to keep them engaged. By accurately gauging their preferences, it is able to serve them with the subset of available content that maximises revenue for the site. Without the constraints of an appropriate regulatory framework, we argue that a…
Study on self-consuming generative models with diverse human curation, focusing on convergence and stability.
problem Analyzing self-consuming generative models with heterogeneous human curation.
method Investigates the asymptotic behavior of retraining dynamics using nonlinear Perron--Frobenius theory and Banach contraction mapping.
result Improves convergence results and provides stability and non-stability analyses for the model.
FinSphere improves stock analysis quality with AI and expert-curated data.
problem Lack of objective evaluation metrics and depth in stock analysis by FinLLMs.
method Developed AnalyScore, curated Stocksis dataset, and FinSphere AI agent.
result FinSphere outperforms general and domain-specific LLMs in generating high-quality stock analysis reports.
Compressed Federated Distillation reduces communication in federated learning.
problem Communication constraints in Federated Learning.
method Compressed Federated Distillation (CFD) leverages soft labels and quantization techniques.
result Reduces communication by more than 4 orders of magnitude compared to Federated Averaging.
SubseasonalClimateUSA dataset improves subseasonal weather forecasting.
problem Challenges in subseasonal weather forecasting, especially skill of physics-based models and integration of local and global variables.
method Curated dataset for training and benchmarking subseasonal forecasting models, including various methods.
result Benchmarking suggests simple and effective ways to improve current operational models.
Improves anomaly detection with contaminated unlabeled data.
problem Weakness in existing semi-supervised anomaly detection methods when unlabeled data contain anomalies.
method Integrates positive-unlabeled learning with deep anomaly detection models.
result Achieves better detection performance on various datasets.
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.
Method curates cost-effective, high-quality datasets using AI models.
problem Costly manual labeling of datasets.
method Probably Approximately Correct Labels (PACL) method.
result Curates high-quality datasets with low overall labeling error.
CrossFilter tackles noisy labels in audio tagging.
problem Noisy labels in large audio datasets.
method CrossFilter framework using multiple representations and multi-task learning.
result Improves audio tagging performance on FSDKaggle2018 and FSDKaggle2019 datasets.
Paper tackles leveraging unlabeled data for PU classification and robust generation.
problem Scarcity of labeled data in machine learning problems.
method Introduces a novel training framework that simultaneously targets PU classification and conditional generation using extra unlabeled data.
result Proves the effectiveness of a Classifier-Noise-Invariant Conditional GAN (CNI-CGAN) that enhances PU classifier performance and leverages extra data.
New methods learn from PU data with non-representative positives.
problem Learning from PU data with non-representative positive classes.
method Integrates negative-unlabeled and unlabeled-unlabeled learning, or uses a recursive risk estimator.
result Effective across various real-world datasets and forms of positive bias.
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.
ORIL learns a reward function from unlabeled data to improve robot learning.
problem Leveraging unlabeled data for robot learning.
method ORIL learns a reward function from demonstrator and unlabeled trajectories, annotates data, and trains an agent via offline reinforcement learning.
result ORIL consistently outperforms BC agents on various robotic tasks.
Enhances adversarial robustness with unlabeled out-of-domain data.
problem Improving robustness of models against adversarial attacks.
method Leveraging unlabeled data from multiple domains to bridge the sample complexity gap in adversarial robustness.
result Better adversarial robustness achieved when unlabeled data comes from a shifted domain.
The article explains how to estimate confusion matrices for classifiers using unlabeled data.
problem Estimating sensitivity and specificity of binary medical diagnostic tests without gold standard tests.
method Modifying diagnostic test solutions to estimate confusion matrices for classifiers on unlabeled data.
result The approach can be used to estimate accuracy statistics for supervised or unsupervised binary classifiers on unlabeled data.
New method learns to weight unlabeled data in semi-supervised learning.
problem Equal weighting of all unlabeled data in semi-supervised learning.
method Adjust weights for each unlabeled example using influence function.
result Technique outperforms state-of-the-art methods on image and language classification tasks.
Bayesian framework uses unlabeled data to improve fairness assessment.
problem Reliable fairness assessment with limited labeled data.
method Hierarchical latent variable model with Bayesian inference.
result Significant reduction in estimation error for fairness metrics.
Paper analyzes how unlabeled data improves SSL and adversarial robustness.
problem Understanding how unlabeled data impacts SSL and adversarial robustness.
method Minimax analysis and adversarial training.
result Reconstruction-based SSL algorithm is rate-optimal under various models and enhances adversarial robustness.
New research shows unlabeled data is equally valuable as labeled data in certain semi-supervised learning scenarios.
problem Improving learning performance with limited labeled data.
method Statistical models with continuous parameters, showing equal utility of unlabeled data under specific conditions.
result The learning rate of semi-supervised learning scales similarly to supervised learning when unlabeled data is abundant.
LoD improves model safety by integrating unlabeled wild data, reducing OOD misclassification.
problem Improving model safety and reliability using unlabeled wild data containing both in-distribution and out-of-distribution samples.
method Intentionally label-noisifying unlabeled wild data to enable joint learning of labeled ID and OOD data, distinguishing losses between ID and OOD samples.
result LoD framework achieves superior OOD detection without requiring thresholds, improving model safety.
New method uses unlabelled data to improve Bayesian Neural Networks.
problem Lack of ability to use unlabelled data in conventional Bayesian Neural Networks.
method Self-supervised Bayesian Neural Networks using contrastive pretraining and variational lower bound optimization.
result Prior predictive distributions capture problem semantics better and improve predictive performance.
Surveying how to use unlabeled data in federated learning.
problem Costly labeling of data limits FL applications.
method Survey and analyze existing research.
result Potential for using unlabeled data in FL.
Paper proposes a new approach to stabilize GAN training by treating generated data as unlabeled.
problem Traditional GAN training treats generated data as negative, ignoring their potential quality.
method Defines positive and unlabeled classification for GANs, treating generated data as unlabeled.
result PUGAN achieves comparable or better performance than sophisticated discriminator stabilization methods.
The paper proposes a SSL framework for complex causal models using unlabelled data.
problem Understanding how unlabelled data can improve SSL in complex causal models.
method The paper explores flexible causal graph structures and designs causal generative models to generate synthetic labelled data.
result The proposed method effectively improves predictive model accuracy using synthetic labelled data generated from unlabelled data.
ORDisCo learns from unlabeled data to improve semi-supervised continual learning.
problem Lack of effective use of unlabeled data in semi-supervised continual learning.
method Deep Online Replay with Discriminator Consistency (ORDisCo) that continually passes the learned data distribution to a classifier and selectively stabilizes discriminator parameters.
result Significant performance improvement on various semi-supervised learning benchmark datasets.
Paper tackles survival data analysis with positive and unlabeled observations.
problem Traditional survival analysis yields biased results with positive-unlabeled data.
method Developed parametric, nonparametric, and machine learning models for positive and unlabeled survival data.
result Proposed estimation method provides valid results for positive-unlabeled survival data.
Cold posteriors in BNNs harm performance, likely due to incorrect likelihood.
problem Cold posteriors in Bayesian neural networks degrade performance.
method Developed a generative model explaining cold posteriors and matched it to the tempered likelihoods.
result Cold posteriors are a result of using the wrong likelihood for image classification datasets.
Improving a semi-supervised image segmentation task has the option of adding more unlabelled images, labelling the unlabelled images or combining both, as neither image acquisition nor expert labelling can be considered trivial in most clinical applications. With a laparoscopic liver image segmentation application, we …
New framework assesses value of labeled vs unlabeled data in latent variable models.
problem Determining the optimal use of labeled and unlabeled data in latent variable models.
method Developed a bias-variance decomposition of the generalization error for method-of-moments latent variable estimation, and introduced a correction for misspecification.
result Labeled data is more valuable than unlabeled data when models are misspecified, but this value can be reduced with correction.
Proposes SSFair to improve fairness in machine learning using unlabeled data.
problem Fairness issues in machine learning systems due to lack of labeled data.
method Semi-supervised learning with neural networks to leverage unlabeled data.
result Improves fairness of decision-making processes without requiring labeled data.
Deep RL detects anomalies from few labeled examples and large unlabeled data.
problem Anomaly detection with limited labeled data and large unlabeled data.
method Deep reinforcement learning to optimize detection of labeled and unlabeled anomalies.
result Significantly outperforms state-of-the-art methods on 48 real-world datasets.
Bayesian analysis shows unlabeled data improve graph-based semi-supervised learning.
problem Improving semi-supervised learning with limited labeled data.
method Bayesian nonparametric approach using unlabeled data for graph-based learning.
result Posterior contracts optimally around the truth with sufficient unlabeled data.
New STKR estimators use unlabeled data for smoother function learning.
problem Leveraging unlabeled data for smoother function learning.
method Spectrally transformed kernel regression (STKR) with scalable implementations.
result STKR can learn any sufficiently smooth function.
SAL framework uses unlabeled data to improve OOD detection.
problem Lack of clean OOD samples makes OOD detection challenging.
method SAL framework separates candidate outliers and trains an OOD classifier.
result SAL achieves state-of-the-art performance on benchmarks.