Early Exit Neural Network

Early-exit neural networks (EENNs) aim to improve the efficiency of deep learning models by allowing predictions to be made at multiple stages of inference, rather than only at the final layer. Research focuses on developing effective decision mechanisms to determine when to exit early, optimizing EENN architectures (including adaptations of ResNets and transformers), and applying them to various tasks and hardware constraints, such as embedded systems and IoT devices. This approach offers significant potential for reducing computational costs and latency in resource-limited environments, impacting applications ranging from image classification and speech recognition to real-time sensor data processing.

Papers

September 9, 2024

Early-exit Convolutional Neural Networks
Edanur Demir, Emre Akbas
Neural Network Convolutional Neural Network Scientific Inference Convolutional Neural Network Architecture Early Exit Neural Network

August 9, 2024

PriPHiT: Privacy-Preserving Hierarchical Training of Deep Neural Networks
Yamin Sepehri, Pedram Pad, Pascal Frossard, L. Andrea Dunbar
Deep Neural Network Privacy Preservation Differential Privacy Guarantee Reconstruction Attack Early Exit Neural Network Hierarchical Training

March 12, 2024

January 24, 2024

NACHOS: Neural Architecture Search for Hardware Constrained Early Exit Neural Networks
Matteo Gambella, Jary Pomponi, Simone Scardapane, Manuel Roveri
Neural Architecture Search Early Exit Neural Network

November 10, 2023

Early-Exit Neural Networks with Nested Prediction Sets
Metod Jazbec, Patrick Forré, Stephan Mandt, Dan Zhang, Eric Nalisnick
Conformal Prediction Prediction Set Sequential Prediction Early Exit Neural Network Confidence Sequence

September 11, 2023

Temporal Patience: Efficient Adaptive Deep Learning for Embedded Radar Data Processing
Max Sponner, Julius Ott, Lorenzo Servadei, Bernd Waschneck, Robert Wille, Akash Kumar
Radar Data Deep Learning Inference Radar Sensor Early Exit Neural Network Adaptive Deep Radar Processing

June 4, 2023

Finding the SWEET Spot: Analysis and Improvement of Adaptive Inference in Low Resource Settings
Daniel Rotem, Michael Hassid, Jonathan Mamou, Roy Schwartz
General Analysis Large Relevance Improvement Inference Cost Baseline Model Adaptive Inference Early Exit Neural Network SWEET Spot Individual Classifier

April 2, 2023

SEENN: Towards Temporal Spiking Early-Exit Neural Networks
Yuhang Li, Tamar Geller, Youngeun Kim, Priyadarshini Panda
Spiking Neural Network Binary Spike SNN Conversion Early Exit Neural Network

December 25, 2022

QuickNets: Saving Training and Preventing Overconfidence in Early-Exit Neural Architectures
Devdhar Patel, Hava Siegelmann
Neural Network Deep Neural Network Faster Training Early Exit Neural Network

October 27, 2022

Class Based Thresholding in Early Exit Semantic Segmentation Networks
Alperen Görmez, Erdem Koyuncu
Semantic Segmentation Model Neural Collapse True Class Automatic Thresholding Early Exit Early Exit Neural Network Masking Rate

September 27, 2022

Fluid Batching: Exit-Aware Preemptive Serving of Early-Exit Neural Networks on Edge NPUs
Alexandros Kouris, Stylianos I. Venieris, Stefanos Laskaridis, Nicholas D. Lane
Device Inference Processing Unit Early Exit Neural Network Preemptive Answer

June 17, 2022

Binary Early-Exit Network for Adaptive Inference on Low-Resource Devices
Aaqib Saeed
Deep Neural Network Binary Neural Network Resource Constrained Device BNN Layer Adaptive Inference Early Exit Neural Network Binary Model

Early Exit Neural Network

Papers

Early-exit Convolutional Neural Networks

PriPHiT: Privacy-Preserving Hierarchical Training of Deep Neural Networks

Conditional computation in neural networks: principles and research trends

Temporal Decisions: Leveraging Temporal Correlation for Efficient Decisions in Early Exit Neural Networks

Efficient Post-Training Augmentation for Adaptive Inference in Heterogeneous and Distributed IoT Environments

NACHOS: Neural Architecture Search for Hardware Constrained Early Exit Neural Networks

Early-Exit Neural Networks with Nested Prediction Sets

Temporal Patience: Efficient Adaptive Deep Learning for Embedded Radar Data Processing

Finding the SWEET Spot: Analysis and Improvement of Adaptive Inference in Low Resource Settings

SEENN: Towards Temporal Spiking Early-Exit Neural Networks

QuickNets: Saving Training and Preventing Overconfidence in Early-Exit Neural Architectures

Class Based Thresholding in Early Exit Semantic Segmentation Networks

Fluid Batching: Exit-Aware Preemptive Serving of Early-Exit Neural Networks on Edge NPUs

Binary Early-Exit Network for Adaptive Inference on Low-Resource Devices