Latest Articles for DNN

DeepSample offers effective techniques for testing Deep Neural Networks efficiently and accurately.

New method enhances DNNs' reliability against small image shifts.

Introducing a flexible memory framework to optimize deep neural network accelerators.

This article covers the use of deep neural networks to predict meson properties.

A quick way to evaluate DNN performance after new training.

A framework that enhances DNN performance in energy-constrained environments.

New techniques enhance the verification of deep neural networks for safer applications.

Researchers enhance measurements of bright stars using deep neural networks.

New methods for approximating complex functions from limited data are emerging.

New method reduces backdoor threats in deep neural networks.

A new benchmark enhances offline multi-objective optimization methods across various fields.

Improving DNN performance through efficient data sampling methods.

A new optimizer enhances efficiency in running deep neural networks on GPUs.

FRED enhances communication for faster and more efficient deep learning training.

New method enhances spiking neural networks' performance in language tasks.

This paper presents a new DNN-based IDS in programmable networks.

This study explores deep learning models for accurate rice mapping in Bhutan.

Combining pruning and quantization streamlines DNN efficiency for smaller devices.

Exploring the optimization of DNNs for devices with limited energy supply.

A new framework tackles fairness in deep neural networks during model pruning.

A study reveals key differences in how humans and AI represent images.

Examining the role of dropout techniques in improving fairness in DNNs.

SPOGA accelerates deep neural networks with improved speed and energy efficiency.

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

Edge computing improves speed and efficiency in mobile applications using AI.

A new method improves deep learning efficiency for edge devices.

A deep look into how we and machines recognize objects.

A new method enhances testing for deep neural networks by focusing on overall performance.

Introducing DUCD, a privacy-focused method to protect DNNs from adversarial attacks.

HYDRA improves deep neural network efficiency for resource-limited edge devices.

HESSO simplifies model compression, making neural networks more efficient without losing performance.

Study shows staining variations hinder deep neural networks in cancer prediction.

This research explores memory reduction methods for training deep neural networks.

Optimizing DNNs with power-of-two quantization for resource-limited devices.

AIMC chips show promise in defending AI from clever attacks.

A new strategy for targeting multiple tasks in deep neural networks.

This study measures uncertainty in model predictions to detect deceptive design patterns.

Discover how B3FA attacks compromise deep neural networks with minimal knowledge.

Learn how to check if neural networks remain accurate with data changes.

New techniques enhance plant disease detection for farmers using drones and AI.