Deep Neural Network

Deep neural networks (DNNs) are complex computational models aiming to mimic the human brain's learning capabilities, primarily focusing on achieving high accuracy and efficiency in various tasks. Current research emphasizes understanding DNN training dynamics, including phenomena like neural collapse and the impact of architectural choices (e.g., convolutional, transformer, and operator networks) and training strategies (e.g., weight decay, knowledge distillation, active learning). This understanding is crucial for improving DNN performance, robustness (including against adversarial attacks and noisy data), and resource efficiency in diverse applications ranging from image recognition and natural language processing to scientific modeling and edge computing.

Papers

June 1, 2024

Length-scale study in deep learning prediction for non-small cell lung cancer brain metastasis
Haowen Zhou, Steven, Lin, Mark Watson, Cory T. Bernadt, Oumeng Zhang, Ramaswamy Govindan, Richard J. Cote, Changhuei Yang
Deep Neural Network Digital Pathology Survival Prediction Microscopic Image Brain Metastasis Tissue Phenotyping
On the Use of Anchoring for Training Vision Models
Vivek Narayanaswamy, Kowshik Thopalli, Rushil Anirudh, Yamen Mubarka, Wesam Sakla, Jayaraman J. Thiagarajan
Deep Neural Network Greater Public Use Vision Model
SAM-VMNet: Deep Neural Networks For Coronary Angiography Vessel Segmentation
Xueying Zeng, Baixiang Huang, Yu Luo, Guangyu Wei, Songyan He, Yushuang Shao
Deep Neural Network Vessel Segmentation Vascular Segmentation Tomography Segmentation Coronary Artery Segmentation Stenosis Detection
Stochastic Resetting Mitigates Latent Gradient Bias of SGD from Label Noise
Youngkyoung Bae, Yeongwoo Song, Hawoong Jeong
Neural Network Deep Neural Network Stochastic Gradient Descent Noisy Label Generalization Performance Stochastic Way Model Overfitting

May 31, 2024

Searching for internal symbols underlying deep learning
Jung H. Lee, Sujith Vijayan
Deep Learning LeArning Abstract Deep Neural Network Code Data Symbol Description
Certifying Global Robustness for Deep Neural Networks
You Li, Guannan Zhao, Shuyu Kong, Yunqi He, Hai Zhou
Deep Neural Network Robust Deep Global Robustness Local Robustness Robustness Measurement

May 30, 2024

subMFL: Compatiple subModel Generation for Federated Learning in Device Heterogenous Environment
Zeyneddin Oz, Ceylan Soygul Oz, Abdollah Malekjafarian, Nima Afraz, Fatemeh Golpayegani
Deep Neural Network Resource Constrained Device Dense Model Sparsity Constraint Submodel Extraction
Accurate and Reliable Predictions with Mutual-Transport Ensemble
Han Liu, Peng Cui, Bingning Wang, Jun Zhu, Xiaolin Hu
Deep Neural Network Optimal Transport Cross Entropy Loss Uncertainty Calibration Reliable Uncertainty Reliable Prediction

Deep Neural Network

Papers

Length-scale study in deep learning prediction for non-small cell lung cancer brain metastasis

On the Use of Anchoring for Training Vision Models

SAM-VMNet: Deep Neural Networks For Coronary Angiography Vessel Segmentation

Stochastic Resetting Mitigates Latent Gradient Bias of SGD from Label Noise

Searching for internal symbols underlying deep learning

Certifying Global Robustness for Deep Neural Networks

subMFL: Compatiple subModel Generation for Federated Learning in Device Heterogenous Environment

Accurate and Reliable Predictions with Mutual-Transport Ensemble

Exploiting Chaotic Dynamics as Deep Neural Networks

State Space Models are Comparable to Transformers in Estimating Functions with Dynamic Smoothness

EntProp: High Entropy Propagation for Improving Accuracy and Robustness

Towards a theory of how the structure of language is acquired by deep neural networks

2BP: 2-Stage Backpropagation

Trustworthy DNN Partition for Blockchain-enabled Digital Twin in Wireless IIoT Networks

Understanding Forgetting in Continual Learning with Linear Regression

How Does Perfect Fitting Affect Representation Learning? On the Training Dynamics of Representations in Deep Neural Networks

Large Deviations of Gaussian Neural Networks with ReLU activation

Front-propagation Algorithm: Explainable AI Technique for Extracting Linear Function Approximations from Neural Networks

Maintaining and Managing Road Quality:Using MLP and DNN

Method and Software Tool for Generating Artificial Databases of Biomedical Images Based on Deep Neural Networks