mlsanityData to 7 Oct 2026

Leveraging the HW/SW Optimizations and Ecosystems that Drive the AI Revolution

arXiv 2208.02808 ↗First posted 4 Aug 2022cs.LG cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

This paper presents a state-of-the-art overview on how to architect, design, and optimize Deep Neural Networks (DNNs) such that performance is improved and accuracy is preserved. The paper covers a set of optimizations that span the entire Machine Learning processing pipeline. We introduce two types of optimizations. The first alters the DNN model and requires NN re-training, while the second does not. We focus on GPU optimizations, but we believe the presented techniques can be used with other AI inference platforms. To demonstrate the DNN model optimizations, we improve one of the most advanced deep network architectures for optical flow, RAFT arXiv:2003.12039, on a popular edge AI inference platform (Nvidia Jetson AGX Xavier).

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 23 of its 38 references that are in the map. First posted on arXiv 4 Aug 2022.

Builds on

23 of 38 references.

References · 23

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  3. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,385
  4. 2015Learning both Weights and Connections for Efficient Neural Networkscited by 1,998
  5. 2019Searching for MobileNetV3cited by 1,984
  6. 2017Mixed Precision Trainingcited by 1,135
  7. 2021Scaling Vision Transformerscited by 993
  8. 2019Once-for-All: Train One Network and Specialize it for Efficient Deploymentcited by 778
  9. 2018Exploring the Limits of Weakly Supervised Pretrainingcited by 604
  10. 2018Albumentations: fast and flexible image augmentationscited by 558
  11. 2016Learning Structured Sparsity in Deep Neural Networkscited by 538
  12. 2021ResNet strikes back: An improved training procedure in timmcited by 365
11 more
  1. 2019Kornia: an Open Source Differentiable Computer Vision Library for PyTorchcited by 231
  2. 2021Accelerating Sparse Deep Neural Networkscited by 213
  3. 2020Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluationcited by 183
  4. 2020LSQ+: Improving low-bit quantization through learnable offsets and better initializationcited by 180
  5. 2020Sparse GPU Kernels for Deep Learningcited by 138
  6. 2021Dive into Deep Learningcited by 111
  7. 2018NVIDIA Tensor Core Programmability, Performance & Precisioncited by 77
  8. 2018The History Began from AlexNet: A Comprehensive Survey on Deep Learning Approachescited by 74
  9. 2018Glow: Graph Lowering Compiler Techniques for Neural Networkscited by 73
  10. 2020Hardware and Software Optimizations for Accelerating Deep Neural Networks: Survey of Current Trends, Challenges, and the Road Aheadcited by 34
  11. 2018Image Classification at Supercomputer Scalecited by 16

Built on by

Not cited yet.