Latest Articles for Data Science

Energy GNNs improve communication and adaptation in multi-agent systems.

Model merging combines different AI models for improved performance across tasks.

A new approach using knowledge enhances efficiency in learning from relational graphs.

A new method uses translation to enhance language model training.

Improve model training for coding tasks through effective data pruning techniques.

Missing data affects model performance and insights derived from machine learning.

Examining the impact of anti-discrimination laws on automated decision systems.

A look at nonparametric regression methods using geometric graphs and Bayesian approaches.

A new model enhances recommendation systems by analyzing user search intentions.

New algorithms improve efficiency in distributed bilevel optimization with communication compression.

HIGHT enhances language models by using hierarchical information from graph data.

This work focuses on better number representation using digit embeddings for improved predictions.

A new method enhances efficiency in low-entropy couplings for large distributions.

Learn how Bayesian learning improves predictions and handles uncertainty in machine learning.

Exploring the role of synthetic data in improving schizophrenia diagnosis.

A look into Normalizing Flows and their role in generative modeling.

This article discusses effective methods for improving uncertainty quantification in complex models using neural networks.

A new model simplifies analysis of multilayer networks with diverse data types.

LEQ improves long-horizon task performance in offline reinforcement learning.

Introducing new algorithms for efficient bilevel optimization in machine learning.

A new method speeds up image generation while maintaining quality.

Discover how genetic algorithms can refine hyperparameter tuning in machine learning models.

Tools like OxonFair help ensure fairness in AI decision-making.

WildDESED improves sound detection systems in noisy home environments.

A new framework improves graph classification by focusing on core features.

Examining vulnerabilities from clean-label backdoor attacks and how generalization bounds can help.

HydraGNN uses machine learning to accelerate materials discovery and modeling.

A novel approach enhancing UDA performance using CLIP and language guidance.

A new method for improving generative models using context effectively.

A flexible method enhances model learning by adjusting sample selection based on performance.

A new approach improves learning efficiency in reinforcement learning environments.

Introducing ENSY to enhance minority travel mode predictions.

Weight clipping enhances model performance in deep learning and reinforcement learning.

This study proposes a framework for understanding hidden concepts in complex data.

A framework to reduce bias in AI language models while maintaining accuracy.

New methods improve model recognition across varied data types.

LLP enables model training using average labels from grouped examples.

A new method improves the selection of data mixtures for language model training.

A new method enhances LoRA's efficiency and effectiveness in machine learning.

Exploring how synthetic data shapes machine learning models and their behavior.