Latest Articles for Neural Networks

ISQuant offers a new approach to quantization for efficient model deployment.

Discover how adaptive dynamic quantization enhances VQ-VAE models for better data representation.

This study examines how neural networks interpret speech using spectrograms.

This study explores how transformers learn from Markov processes through initialization and gradient flow.

This study improves transfer learning by optimizing learning rates for each layer.

This study explores the role of feed-forward layers in code language models.

Combining sound and images for smarter recognition systems.

Exploring how neural networks use their learned weights effectively.

This article outlines a new approach using Test-Time Training for enhancing RNN performance.

A method to enhance model efficiency in machine learning through effective pruning strategies.

LayerShuffle enhances the robustness of neural networks by enabling flexible layer execution.

Exploring how Hopfield networks mimic brain memory storage and retrieval.

Introducing a new method for Bayesian neural networks that improves uncertainty modeling.

Exploring fKANs and their impact on machine learning performance.

Study on the influence of receptive field size in U-Net models for image segmentation.

ElasticAST allows processing of variable length audio efficiently without losing important details.

A novel method simplifies complex 3D shapes with effective sweep surfaces.

This article investigates how neural networks process data through their representations.

A new approach enhances CNN training timing and efficiency.

Introducing a method that enhances learning from limited data without forgetting past knowledge.

A look into enhancing FPGA use in DNN applications with new techniques.

Introducing Group-and-Shuffle matrices for efficient fine-tuning of neural models.

A study on enhancing decision-making in limited-information chess through neural networks.

A new method using circular vectors improves efficiency in multi-label tasks.

LeRF combines deep learning and interpolation for better image resizing.

This article examines how Transformers reason and the role of scratchpads.

A novel method addresses key challenges in reinforcement learning through improved optimization techniques.

Examining the impact of periodic activation functions on learning efficiency and generalization.

CCL ensures neural networks maintain accuracy while learning new tasks.

Machine learning enhances quantum control techniques for improved technology applications.

Using neural networks to identify chiral magnetic waves in particle physics.

MambaVision combines Mamba and Transformers for better image recognition.

Study reveals how sparsity in AI models changes across layers during training.

DisMAE enhances model generalization across domains using unlabeled data.

A fresh approach to improve gamma-ray observations using neural networks.

A hybrid model improves image restoration using Spiking and Convolutional Neural Networks.

This article discusses new methods improving deep learning performance using nonlocal derivatives.

This article examines Adagrad's efficacy and its advantages over standard methods in large batch training.

A study on using neural networks for simulating material phase dynamics.

A study on machine learning techniques for modeling atomic systems.