DeepAI AI Chat
Log In Sign Up

Improving the Accuracy of Early Exits in Multi-Exit Architectures via Curriculum Learning

by   Arian Bakhtiarnia, et al.

Deploying deep learning services for time-sensitive and resource-constrained settings such as IoT using edge computing systems is a challenging task that requires dynamic adjustment of inference time. Multi-exit architectures allow deep neural networks to terminate their execution early in order to adhere to tight deadlines at the cost of accuracy. To mitigate this cost, in this paper we introduce a novel method called Multi-Exit Curriculum Learning that utilizes curriculum learning, a training strategy for neural networks that imitates human learning by sorting the training samples based on their difficulty and gradually introducing them to the network. Experiments on CIFAR-10 and CIFAR-100 datasets and various configurations of multi-exit architectures show that our method consistently improves the accuracy of early exits compared to the standard training approach.


Single-Layer Vision Transformers for More Accurate Early Exits with Less Overhead

Deploying deep learning models in time-critical applications with limite...

ScreenerNet: Learning Curriculum for Neural Networks

We propose to learn a curriculum or a syllabus for supervised learning w...

LeRaC: Learning Rate Curriculum

Most curriculum learning methods require an approach to sort the data sa...

Multi-Exit Vision Transformer for Dynamic Inference

Deep neural networks can be converted to multi-exit architectures by ins...

Rethinking Curriculum Learning with Incremental Labels and Adaptive Compensation

Like humans, deep networks learn better when samples are organized and i...

A Study on the Predictability of Sample Learning Consistency

Curriculum Learning is a powerful training method that allows for faster...

Conflicting Bundles: Adapting Architectures Towards the Improved Training of Deep Neural Networks

Designing neural network architectures is a challenging task and knowing...