# Training BatchNorm and Only BatchNorm: On the Expressive Power of Random Features in CNNs

arXiv: https://arxiv.org/abs/2003.00152 · HTML: https://mlsanity.com/paper/2003.00152/

## Key facts

- Cited by 104 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Feb 2020.

## Built on by (most cited first, 30 of 104)

- [BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-models](https://mlsanity.com/paper/2106.10199/) (2021)
- [VeRA: Vector-based Random Matrix Adaptation](https://mlsanity.com/paper/2310.11454/) (2023)
- [Point-Bind & Point-LLM: Aligning Point Cloud with Multi-modality for 3D Understanding, Generation, and Instruction Following](https://mlsanity.com/paper/2309.00615/) (2023)
- [Masked Language Modeling and the Distributional Hypothesis: Order Word Matters Pre-training for Little](https://mlsanity.com/paper/2104.06644/) (2021)
- [Pretrained Transformers as Universal Computation Engines](https://mlsanity.com/paper/2103.05247/) (2021)
- [On-Device Training Under 256KB Memory](https://mlsanity.com/paper/2206.15472/) (2022)
- [Efficient Self-supervised Vision Transformers for Representation Learning](https://mlsanity.com/paper/2106.09785/) (2021)
- [ImageBind-LLM: Multi-modality Instruction Tuning](https://mlsanity.com/paper/2309.03905/) (2023)
- [MultiEURLEX -- A multi-lingual and multi-label legal document classification dataset for zero-shot cross-lingual transfer](https://mlsanity.com/paper/2109.00904/) (2021)
- [CLIP for All Things Zero-Shot Sketch-Based Image Retrieval, Fine-Grained or Not](https://mlsanity.com/paper/2303.13440/) (2023)
- [LPT: Long-tailed Prompt Tuning for Image Classification](https://mlsanity.com/paper/2210.01033/) (2022)
- [Parameter-Efficient Fine-Tuning for Medical Image Analysis: The Missed Opportunity](https://mlsanity.com/paper/2305.08252/) (2023)
- [Multi-Prize Lottery Ticket Hypothesis: Finding Accurate Binary Neural Networks by Pruning A Randomly Weighted Network](https://mlsanity.com/paper/2103.09377/) (2021)
- [Towards In-context Scene Understanding](https://mlsanity.com/paper/2306.01667/) (2023)
- [Tiny Machine Learning: Progress and Futures](https://mlsanity.com/paper/2403.19076/) (2024)
- [Enable Deep Learning on Mobile Devices: Methods, Systems, and Applications](https://mlsanity.com/paper/2204.11786/) (2022)
- [Batch Normalization Orthogonalizes Representations in Deep Random Networks](https://mlsanity.com/paper/2106.03970/) (2021)
- [Normalization Layers Are All That Sharpness-Aware Minimization Needs](https://mlsanity.com/paper/2306.04226/) (2023)
- [Single-Layer Vision Transformers for More Accurate Early Exits with Less Overhead](https://mlsanity.com/paper/2105.09121/) (2021)
- [AlphaTuning: Quantization-Aware Parameter-Efficient Adaptation of Large-Scale Pre-Trained Language Models](https://mlsanity.com/paper/2210.03858/) (2022)
- [Learning to Generate Image Embeddings with User-level Differential Privacy](https://mlsanity.com/paper/2211.10844/) (2022)
- [PockEngine: Sparse and Efficient Fine-tuning in a Pocket](https://mlsanity.com/paper/2310.17752/) (2023)
- [BN-NAS: Neural Architecture Search with Batch Normalization](https://mlsanity.com/paper/2108.07375/) (2021)
- [Unsupervised Image Representation Learning with Deep Latent Particles](https://mlsanity.com/paper/2205.15821/) (2022)
- [PECoP: Parameter Efficient Continual Pretraining for Action Quality Assessment](https://mlsanity.com/paper/2311.07603/) (2023)
- [TWINS: A Fine-Tuning Framework for Improved Transferability of Adversarial Robustness and Generalization](https://mlsanity.com/paper/2303.11135/) (2023)
- [Sparse Structure Search for Parameter-Efficient Tuning](https://mlsanity.com/paper/2206.07382/) (2022)
- [Towards Foundation Models and Few-Shot Parameter-Efficient Fine-Tuning for Volumetric Organ Segmentation](https://mlsanity.com/paper/2303.17051/) (2023)
- [Modality-Agnostic Structural Image Representation Learning for Deformable Multi-Modality Medical Image Registration](https://mlsanity.com/paper/2402.18933/) (2024)
- [Partial transfusion: on the expressive influence of trainable batch norm parameters for transfer learning](https://mlsanity.com/paper/2102.05543/) (2021)

## Abstract (arXiv)

A wide variety of deep learning techniques from style transfer to multitask learning rely on training affine transformations of features. Most prominent among these is the popular feature normalization technique BatchNorm, which normalizes activations and then subsequently applies a learned affine transform. In this paper, we aim to understand the role and expressive power of affine parameters used to transform features in this way. To isolate the contribution of these parameters from that of the learned features they transform, we investigate the performance achieved when training only these parameters in BatchNorm and freezing all weights at their random initializations. Doing so leads to surprisingly high performance considering the significant limitations that this style of training imposes. For example, sufficiently deep ResNets reach 82% (CIFAR-10) and 32% (ImageNet, top-5) accuracy in this configuration, far higher than when training an equivalent number of randomly chosen parameters elsewhere in the network. BatchNorm achieves this performance in part by naturally learning to disable around a third of the random features. Not only do these results highlight the expressive power of affine parameters in deep learning, but - in a broader sense - they characterize the expressive power of neural networks constructed simply by shifting and rescaling random features.