mlsanityData to 7 Oct 2026

Tailor: Altering Skip Connections for Resource-Efficient Inference

arXiv 2301.07247 ↗First posted 18 Jan 2023cs.CV cs.LG cs.NEcited by 6

Not on the map. Why?

Abstract (arXiv)

Deep neural networks use skip connections to improve training convergence. However, these skip connections are costly in hardware, requiring extra buffers and increasing on- and off-chip memory utilization and bandwidth requirements. In this paper, we show that skip connections can be optimized for hardware when tackled with a hardware-software codesign approach. We argue that while a network's skip connections are needed for the network to learn, they can later be removed or shortened to provide a more hardware efficient implementation with minimal to no accuracy loss. We introduce Tailor, a codesign tool whose hardware-aware training algorithm gradually removes or shortens a fully trained network's skip connections to lower their hardware cost. Tailor improves resource utilization by up to 34% for BRAMs, 13% for FFs, and 16% for LUTs for on-chip, dataflow-style architectures. Tailor increases performance by 30% and reduces memory bandwidth by 45% for a 2D processing element array architecture.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 30 of its 54 references that are in the map. First posted on arXiv 18 Jan 2023.

Builds on

30 of 54 references.

References · 30

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  4. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  5. 2016Densely Connected Convolutional Networkscited by 6,691
  6. 2016Identity Mappings in Deep Residual Networkscited by 1,960
  7. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,548
  8. 2019A Comprehensive Survey on Transfer Learningcited by 1,154
  9. 2018ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Designcited by 1,132
  10. 2017Visualizing the Loss Landscape of Neural Netscited by 1,126
  11. 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 798
  12. 2019Improved Knowledge Distillation via Teacher Assistantcited by 567
18 more
  1. 2021RepVGG: Making VGG-style ConvNets Great Againcited by 520
  2. 2018HAQ: Hardware-Aware Automated Quantization with Mixed Precisioncited by 426
  3. 2019HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precisioncited by 387
  4. 2019PyHessian: Neural Networks Through the Lens of the Hessiancited by 291
  5. 2019HAWQ-V2: Hessian Aware trace-Weighted Quantization of Neural Networkscited by 250
  6. 2018Fast inference of deep neural networks in FPGAs for particle physicscited by 132
  7. 2021MLPerf Tiny Benchmarkcited by 118
  8. 2019QuartzNet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutionscited by 93
  9. 2018ResNet with one-neuron hidden layers is a Universal Approximatorcited by 78
  10. 2021hls4ml: An Open-Source Codesign Workflow to Empower Scientific Low-Power Machine Learning Devicescited by 61
  11. 2017DiracNets: Training Very Deep Neural Networks Without Skip-Connectionscited by 47
  12. 2021Fast convolutional neural networks on FPGAs with hls4mlcited by 42
  13. 2018Equivalence of approximation by convolutional neural networks and fully-connected networkscited by 38
  14. 2016Deep Model Compression: Distilling Knowledge from Noisy Teacherscited by 35
  15. 2020Mix and Match: A Novel FPGA-Centric Deep Neural Network Quantization Frameworkcited by 27
  16. 2017Minimum Energy Quantized Neural Networkscited by 20
  17. 2018Synetgy: Algorithm-hardware Co-design for ConvNet Accelerators on Embedded FPGAscited by 18
  18. 2022Open-source FPGA-ML codesign for the MLPerf Tiny Benchmarkcited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 5; 2025: 0; 2026: 1.

6 papers.

  1. 2024A Lightweight Spatiotemporal Network for Online Eye Tracking with Event Cameracited by 14
  2. 2024Towards Narrowing the Generalization Gap in Deep Boolean Networkscited by 3
  3. 2026SwiftChannel: Algorithm-Hardware Co-Design for Deep Learning-Based 5G Channel Estimationcited by 1
  4. 2024AdaResNet: Enhancing Residual Networks with Dynamic Weight Adjustment for Improved Feature Integrationcited by 0
  5. 2024Error Diffusion: Post Training Quantization with Block-Scaled Number Formats for Neural Networkscited by 0
  6. 2024UNet--: Memory-Efficient and Feature-Enhanced Network Architecture based on U-Net with Reduced Skip-Connectionscited by 0