Neural Network

Neural networks are computational models inspired by the structure and function of the brain, primarily aimed at approximating complex functions and solving diverse problems through learning from data. Current research emphasizes improving efficiency and robustness, exploring novel architectures like sinusoidal neural fields and hybrid models combining neural networks with radial basis functions, as well as developing methods for understanding and manipulating the internal representations learned by these networks, such as through hyper-representations of network weights. These advancements are driving progress in various fields, including computer vision, natural language processing, and scientific modeling, by enabling more accurate, efficient, and interpretable AI systems.

Papers

October 15, 2024

DiffGAN: A Test Generation Approach for Differential Testing of Deep Neural Networks
Zohreh Aghababaeyan, Manel Abdellatif, Lionel Briand, Ramesh S
Neural Network Deep Neural Network Adversarial Learning DNN Model Test Generation Differential Testing DNN Prediction
G-Designer: Architecting Multi-agent Communication Topologies via Graph Neural Networks
Guibin Zhang, Yanwei Yue, Xiangguo Sun, Guancheng Wan, Miao Yu, Junfeng Fang, Kun Wang, Dawei Cheng
Neural Network Graph Drawing Multi Agent Network Topology Agent Communication Multi Agent Communication Virtual Network
Degradation Oriented and Regularized Network for Real-World Depth Super-Resolution
Zhengxue Wang, Zhiqiang Yan
Neural Network Super Resolution Degradation Representation Document Degradation
Are High-Degree Representations Really Unnecessary in Equivariant Graph Neural Networks?
Jiacheng Cen, Anyi Li, Ning Lin, Yuxiang Ren, Zihe Wang, Wenbing Huang
Neural Network Level Representation Equivariant Graph Neural Network Equivariant Graph Steering Vector
Error Diffusion: Post Training Quantization with Block-Scaled Number Formats for Neural Networks
Alireza Khodamoradi, Kristof Denolf, Eric Dellinger
Neural Network Post Training Quantization Low Precision Quantization Error Textual Input Error Metric

October 14, 2024

Statistical Properties of Deep Neural Networks with Dependent Data
Chad Brown
Neural Network Deep Neural Network Dependent Data Statistical Property Deep Neural Network Estimator
Towards a More Complete Theory of Function Preserving Transforms
Michael Painter
Neural Network Deep Learning Architecture Residual Connection Preserving Transformation
Hard-Constrained Neural Networks with Universal Approximation Guarantees
Youngjae Min, Anoopkumar Sonar, Navid Azizan
Neural Network Constraint Satisfaction Hard Constraint Unconstrained Optimization Convex Constraint Soft Constraint
Dynamical loss functions shape landscape topography and improve learning in artificial neural networks
Eduardo Lavin, Miguel Ruiz-Garcia
Neural Network Loss Function Loss Landscape Dynamic Loss
Neural networks that overcome classic challenges through practice
Kazuki Irie, Brenden M. Lake
Neural Network LeArning Abstract Strong Generalization Practice Mode Neural Network Model Open Challenge Cognitive Capability Metalearning Algorithm
Feature Averaging: An Implicit Bias of Gradient Descent Leading to Non-Robustness in Neural Networks
Binghui Li, Zhixuan Pan, Kaifeng Lyu, Jian Li
Neural Network Native Robustness Gradient Descent Stochastic Gradient Descent Implicit Bias Non Robust
Fast and Accurate Neural Rendering Using Semi-Gradients
In-Young Cho, Jaewoong Cho
Neural Network Real Time Rendering Free Viewpoint Rendering Illumination Model Well Learned Radiance

October 13, 2024

EMWaveNet: Physically Explainable Neural Network Based on Microwave Propagation for SAR Target Recognition
Zhuoxuan Li, Xu Zhang, Shumeng Yu, Haipeng Wang
Neural Network Synthetic Aperture Radar Fast Propagation Target Recognition
Gradient-Free Neural Network Training on the Edge
Dotan Di Castro, Omkar Joglekar, Shir Kozlovsky, Vladimir Tchuiev, Michal Moshkovitz
Neural Network Extreme Edge Gradient Based Low Precision

October 12, 2024

DuoDiff: Accelerating Diffusion Models with a Dual-Backbone Approach
Daniel Gallo Fernández, Rǎzvan-Andrei Matişan, Alejandro Monroy Muñoz, Ana-Maria Vasilcoiu, Janusz Partyka, Tin Hadži Veljković, Metod Jazbec
Neural Network Diffusion Model Denoising Network Multi Exit Backbone Architecture
Structure of Artificial Neural Networks -- Empirical Investigations
Julian Stier
Neural Network Neural Architecture Search Empirical Study Neural Architecture Inner Structure
PrivQuant: Communication-Efficient Private Inference with Quantized Network/Protocol Co-Optimization
Tianshi Xu, Shuzhang Zhong, Wenxuan Zeng, Runsheng Wang, Meng Li
Neural Network Network Programming Inference Strategy Private Inference Co Optimization

Neural Network

Papers

Loss Landscape Characterization of Neural Networks without Over-Parametrziation

Stable Diffusion with Continuous-time Neural Network

LPUF-AuthNet: A Lightweight PUF-Based IoT Authentication via Tandem Neural Networks and Split Learning

DiffGAN: A Test Generation Approach for Differential Testing of Deep Neural Networks

G-Designer: Architecting Multi-agent Communication Topologies via Graph Neural Networks

Degradation Oriented and Regularized Network for Real-World Depth Super-Resolution

Are High-Degree Representations Really Unnecessary in Equivariant Graph Neural Networks?

Error Diffusion: Post Training Quantization with Block-Scaled Number Formats for Neural Networks

Statistical Properties of Deep Neural Networks with Dependent Data

Towards a More Complete Theory of Function Preserving Transforms

Hard-Constrained Neural Networks with Universal Approximation Guarantees

Dynamical loss functions shape landscape topography and improve learning in artificial neural networks

Neural networks that overcome classic challenges through practice

Feature Averaging: An Implicit Bias of Gradient Descent Leading to Non-Robustness in Neural Networks

Fast and Accurate Neural Rendering Using Semi-Gradients

EMWaveNet: Physically Explainable Neural Network Based on Microwave Propagation for SAR Target Recognition

Gradient-Free Neural Network Training on the Edge

DuoDiff: Accelerating Diffusion Models with a Dual-Backbone Approach

Structure of Artificial Neural Networks -- Empirical Investigations

PrivQuant: Communication-Efficient Private Inference with Quantized Network/Protocol Co-Optimization