Abstract:Abstract Most curriculum learning methods require an approach to sort the data samples by difficulty, which is often cumbersome to perform. In this work, we propose a novel curriculum learning approach termed Learning Rate Curriculum (LeRaC), which leverages the use of a different learning rate for each layer of a neural network to create a data-agnostic curriculum during the initial training epochs. More specifically, LeRaC assigns higher learning rates to neural layers closer to the input, gradually decreasing the learning rates as the layers are placed farther away from the input. The learning rates increase at various paces during the first training iterations, until they all reach the same value. From this point on, the neural model is trained as usual. This creates a model-level curriculum learning strategy that does not require sorting the examples by difficulty and is compatible with any neural network, generating higher performance levels regardless of the architecture. We conduct comprehensive experiments on 12 data sets from the computer vision (CIFAR-10, CIFAR-100, Tiny ImageNet, ImageNet-1K, Food-101, UTKFace, PASCAL VOC), language (BoolQ, QNLI, RTE) and audio (ESC-50, CREMA-D) domains, considering various convolutional (ResNet-18, Wide-ResNet-50, DenseNet-121, YOLOv5), recurrent (LSTM) and transformer (CvT, BERT, SepTr) architectures. We compare our approach with the conventional training regime, as well as with Curriculum by Smoothing (CBS), a state-of-the-art data-agnostic curriculum learning approach. Unlike CBS, our performance improvements over the standard training regime are consistent across all data sets and models. Furthermore, we significantly surpass CBS in terms of training time (there is no additional cost over the standard training regime for LeRaC). Our code is freely available at: https://github.com/CroitoruAlin/LeRaC .

Cup Curriculum: Curriculum Learning on Model Capacity

A Comprehensive Survey on Curriculum Learning

A Survey on Curriculum Learning

Curriculum learning for language modeling

Curriculum Learning with Adam: The Devil Is in the Wrong Details

Curriculum Learning: A Survey

An Analytical Theory of Curriculum Learning in Teacher-Student Networks

Irreducible Curriculum for Language Model Pretraining

A Psychology-based Unified Dynamic Framework for Curriculum Learning

Review and Arrange: Curriculum Learning for Natural Language Understanding

Let's Learn Step by Step: Enhancing In-Context Learning Ability with Curriculum Learning

Dynamic curriculum learning for conversation response selection

CLIMB: Curriculum Learning for Infant-inspired Model Building

Coarse-to-Fine Curriculum Learning

Visualizing and Understanding Curriculum Learning for Long Short-Term Memory Networks

Cyclical Curriculum Learning

Curriculumformer: Taming Curriculum Pre-Training for Enhanced 3-D Point Cloud Understanding

Learning to Learn: How to Continuously Teach Humans and Machines

Curriculum Learning for Small Code Language Models

Learning Rate Curriculum

Exploring Curriculum Learning for Vision-Language Tasks: A Study on Small-Scale Multimodal Training