Revisiting ResNets: Improved Training and Scaling Strategies

Abstract

Novel computer vision architectures monopolize the spotlight, but the impact of the model architecture is often conflated with simultaneous changes to training methodology and scaling strategies. Our work revisits the canonical ResNet (He et al., 2015) and studies these three aspects in an effort to disentangle them. Perhaps surprisingly, we find that training and scaling strategies may matter more than architectural changes, and further, that the resulting ResNets match recent state-of-the-art models. We show that the best performing scaling strategy depends on the training regime and offer two new scaling strategies: (1) scale model depth in regimes where overfitting can occur (width scaling is preferable otherwise); (2) increase image resolution more slowly than previously recommended (Tan & Le, 2019). Using improved training and scaling strategies, we design a family of ResNet architectures, ResNet-RS, which are 1.7x - 2.7x faster than EfficientNets on TPUs, while achieving similar accuracies on ImageNet. In a large-scale semi-supervised learning setup, ResNet-RS achieves 86.2% top-1 ImageNet accuracy, while being 4.7x faster than EfficientNet NoisyStudent. The training techniques improve transfer performance on a suite of downstream tasks (rivaling state-of-the-art self-supervised algorithms) and extend to video classification on Kinetics-400. We recommend practitioners use these simple revised ResNets as baselines for future research.

References in corpus (9)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Language Models are Few-Shot Learners
• The Kinetics Human Action Video Dataset
• Scaling Laws for Neural Language Models
• Deep Learning Scaling is Predictable, Empirically
• High-Performance Large-Scale Image Recognition Without Normalization
• Stand-Alone Self-Attention in Vision Models
• Scaling Laws for Autoregressive Generative Modeling
• LambdaNetworks: Modeling Long-Range Interactions Without Attention

Cited by in corpus (9)

• MLP-Mixer: An all-MLP Architecture for Vision
• Exploring the Limits of Large Scale Pre-training
• ResNet strikes back: An improved training procedure in timm
• Revisiting 3D ResNets for Video Recognition
• Recurrence along Depth: Deep Convolutional Neural Networks with Recurrent Layer Aggregation
• The Benchmark Lottery
• Making EfficientNet More Efficient: Exploring Batch-Independent Normalization, Group Convolutions and Reduced Resolution Training
• UniNet: Unified Architecture Search with Convolution, Transformer, and MLP
• Analyze and Design Network Architectures by Recursion Formulas