# A Systematic Approach to Blocking Convolutional Neural Networks

arXiv: https://arxiv.org/abs/1606.04209 · HTML: https://mlsanity.com/paper/1606.04209/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Jun 2016.

## Built on by (most cited first, 7 of 7)

- [HW-NAS-Bench:Hardware-Aware Neural Architecture Search Benchmark](https://mlsanity.com/paper/2103.10584/) (2021)
- [Auto-NBA: Efficient and Effective Search Over the Joint Space of Networks, Bitwidths, and Accelerators](https://mlsanity.com/paper/2106.06575/) (2021)
- [Neural Architecture Search as Program Transformation Exploration](https://mlsanity.com/paper/2102.06599/) (2021)
- [RT-RCG: Neural Network and Accelerator Search Towards Effective and Real-time ECG Reconstruction from Intracardiac Electrograms](https://mlsanity.com/paper/2111.02569/) (2021)
- [A Construction Kit for Efficient Low Power Neural Network Accelerator Designs](https://mlsanity.com/paper/2106.12810/) (2021)
- [AMULET: Adaptive Matrix-Multiplication-Like Tasks](https://mlsanity.com/paper/2305.08872/) (2023)
- [KAPLA: Pragmatic Representation and Fast Solving of Scalable NN Accelerator Dataflow](https://mlsanity.com/paper/2306.15676/) (2023)

## Abstract (arXiv)

Convolutional Neural Networks (CNNs) are the state of the art solution for many computer vision problems, and many researchers have explored optimized implementations. Most implementations heuristically block the computation to deal with the large data sizes and high data reuse of CNNs. This paper explores how to block CNN computations for memory locality by creating an analytical model for CNN-like loop nests. Using this model we automatically derive optimized blockings for common networks that improve the energy efficiency of custom hardware implementations by up to an order of magnitude. Compared to traditional CNN CPU implementations based on highly-tuned, hand-optimized BLAS libraries,our x86 programs implementing the optimal blocking reduce the number of memory accesses by up to 90%.