Latest Articles for Neural Networks

A new approach to offline reinforcement learning improves policy learning using diffusion models.

A new approach for generating programs based on images using advanced neural models.

A new approach to improve efficiency in neural architecture search processes.

Research on optimizing deep learning models with sparsity and quantization techniques.

This study investigates how small changes can mislead CNNs in critical tasks.

Exploring advanced methods for effective graph data analysis.

New model improves long-range information flow in graph data.

MaxLin improves CNN verification accuracy and efficiency for safer AI applications.

A new method improves efficient deep learning models through exact orthogonality.

A new weight decay method enhances sparsification in neural networks.

A framework to enhance neural networks by integrating human knowledge into learning algorithms.

New methods reveal resilience in neural network circuits against manipulation.

New methods enhance main task performance using auxiliary data without extra computation costs.

This article examines layer normalization's role in improving neural network classification.

This study explores advanced methods for efficient data labeling using neural network techniques.

This article examines how ReLU networks approximate low regularity functions.

DSNNs process information like real neurons, offering improved efficiency for data handling.

New methods promise faster, efficient neural networks with less resource use.

A method to enhance decision-making in reinforcement learning using representation learning.

This article examines how noise can improve machine learning model performance during training.

CADE optimizes spiking neural networks for better performance and efficiency.

A new method combines deep learning with polynomial techniques for improved function approximations.

Discover how Extended Mind Transformers improve memory handling in language models.

This study highlights the significance of the Neural Tangent Kernel in training neural networks.

This article examines how planning budgets affect DNC models in solving problems.

Exploring how LLMs use reasoning to tackle complex tasks.

A new method enhances GNN training efficiency using Direct Feedback Alignment.

A new method enhances decision-making in reinforcement learning through action-conditional predictions.

A new method for better insights into RNN training dynamics.

This article discusses methods for verifying neural networks in reach-avoid tasks.

Exploring the connections and functions of neurons in processing information.

Study reveals how groups of neurons interact in unique configurations.

A new approach enhances SNNs by converting ANNs effectively.

Tackling the issues of OOD generalization and feature contamination in AI models.

HesScale improves efficiency in machine learning by estimating the Hessian diagonal.

A new framework combining TNNs and persistent homology for better data analysis.

A novel approach to integrate transformers with graph structures for better outcomes.

A new S6 model improves performance and efficiency in spiking neural networks.

Examining the role of neurons in CLIP models and their interactions.

An analysis of Transformers' struggles with counting and copying tasks.