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.

2129papers

Papers - Page 60

June 5, 2023

Adversarial Ink: Componentwise Backward Error Attacks on Deep Learning
Lucas Beerens, Desmond J. Higham
Adversarial Learning Adversarial Attack Deep Learning Deep Neural Network Attack Algorithm
Towards Better Explanations for Object Detection
Van Binh Truong, Truong Thanh Hung Nguyen, Vo Thanh Khang Nguyen, Quoc Khanh Nguyen, Quoc Hung Cao
Deep Neural Network Artificial Intelligence Object Detection Model Better Explainability

June 2, 2023

Overcoming the Stability Gap in Continual Learning
Md Yousuf Harun, Christopher Kanan
Continual LEArning Deep Neural Network Model Collapse Incremental Learning
Exploring Robustness of Image Recognition Models on Hardware Accelerators
Nikolaos Louloudakis, Perry Gibson, José Cano, Ajitha Rajan
Recognition Model Deep Neural Network Hardware Accelerator Mutation Testing
Uniform Convergence of Deep Neural Networks with Lipschitz Continuous Activation Functions and Variable Widths
Yuesheng Xu, Haizhang Zhang
Convergence Analysis Deep Neural Network Lipschitz Continuity Assumption Uniform Convergence Lipschitz Activation Weak Convergence
ErfReLU: Adaptive Activation Function for Deep Neural Network
Ashish Rajanand, Pradeep Singh
Activation Function Adaptive Activation Function Deep Neural Network
Network Degeneracy as an Indicator of Training Performance: Comparing Finite and Infinite Width Angle Predictions
Cameron Jakub, Mihai Nica
Width Network Deep Neural Network Finite Data Large Depth Efficient Training Homogeneous Diffusive Network Neural Network ReLU Layer Aggregate Level Indicator
DVFO: Learning-Based DVFS for Energy-Efficient Edge-Cloud Collaborative Inference
Ziyang Zhang, Yang Zhao, Huan Li, Changyao Lin, Jie Liu
Deep Neural Network Collaborative Inference

June 1, 2023

Overview of Deep Learning Methods for Retinal Vessel Segmentation
Gorana Gojić, Ognjen Kundačina, Dragiša Mišković, Dinu Dragan
Deep Neural Network Retinal Vessel Segmentation Deep Learning Recent Deep Learning
Versatile Backdoor Attack with Visible, Semantic, Sample-Specific, and Compatible Triggers
Ruotong Wang, Hongrui Chen, Zihao Zhu, Li Liu, Baoyuan Wu
Trigger Point Backdoor Attack Trigger Design Deep Neural Network Natural Trigger
Initial Guessing Bias: How Untrained Networks Favor Some Classes
Emanuele Francazi, Aurelien Lucchi, Marco Baity-Jesi
Deep Neural Network Visual Based Class Untrained Neural Network Neural Network Self Averaging
A New PHO-rmula for Improved Performance of Semi-Structured Networks
David Rügamer
Deep Neural Network Structured Network Post Hoc Orthogonalization Structured Regression Performance Improvement
(Almost) Provable Error Bounds Under Distribution Shift via Disagreement Discrepancy
Elan Rosenfeld, Saurabh Garg
Task Discrepancy Deep Neural Network Disagreement Problem Proxy Based Loss Distribution Shift Error Bound

May 31, 2023

Inconsistency, Instability, and Generalization Gap of Deep Neural Network Training
Rie Johnson, Tong Zhang
Deep Neural Network Training Deep Hard to Easy Inconsistency Training Data Core Stability Generalization Gap
Special Session: Approximation and Fault Resiliency of DNN Accelerators
Mohammad Hasan Ahmadilivani, Mario Barbareschi, Salvatore Barone, Alberto Bosio, Masoud Daneshtalab, Salvatore Della Torca, Gabriele Gavarini+5
DNN Accelerator Average Approximation Special Session Fault Resilience Deep Learning Deep Neural Network
Power Control with QoS Guarantees: A Differentiable Projection-based Unsupervised Learning Framework
Mehrazin Alizadeh, Hina Tabassum
Differentiable Projection Power Control Differentiable Convex Interference Channel Deep Neural Network Resource Allocation Quality of Service
The Tunnel Effect: Building Data Representations in Deep Neural Networks
Wojciech Masarczyk, Mateusz Ostaszewski, Ehsan Imani, Razvan Pascanu, Piotr Miłoś, Tomasz Trzciński
Deep Neural Network Continual LEArning Deeper Network Deep Network Separable Representation

Deep Neural Network

Papers - Page 60

Adversarial Ink: Componentwise Backward Error Attacks on Deep Learning

Towards Better Explanations for Object Detection

Overcoming the Stability Gap in Continual Learning

Exploring Robustness of Image Recognition Models on Hardware Accelerators

Uniform Convergence of Deep Neural Networks with Lipschitz Continuous Activation Functions and Variable Widths

ErfReLU: Adaptive Activation Function for Deep Neural Network

Network Degeneracy as an Indicator of Training Performance: Comparing Finite and Infinite Width Angle Predictions

DVFO: Learning-Based DVFS for Energy-Efficient Edge-Cloud Collaborative Inference

Overview of Deep Learning Methods for Retinal Vessel Segmentation

Versatile Backdoor Attack with Visible, Semantic, Sample-Specific, and Compatible Triggers

Initial Guessing Bias: How Untrained Networks Favor Some Classes

A New PHO-rmula for Improved Performance of Semi-Structured Networks

(Almost) Provable Error Bounds Under Distribution Shift via Disagreement Discrepancy

Inconsistency, Instability, and Generalization Gap of Deep Neural Network Training

Special Session: Approximation and Fault Resiliency of DNN Accelerators

Power Control with QoS Guarantees: A Differentiable Projection-based Unsupervised Learning Framework

The Tunnel Effect: Building Data Representations in Deep Neural Networks

Benign Overfitting in Deep Neural Networks under Lazy Training

What Can We Learn from Unlearnable Datasets?

Improving Generalization of Complex Models under Unbounded Loss Using PAC-Bayes Bounds