A Distributed Data-Parallel PyTorch Implementation of the Distributed Shampoo Optimizer for Training Neural Networks At-Scale

• URL: http://arxiv.org/abs/2309.06497v1
• Date: Tue, 12 Sep 2023 18:11:10 GMT
• Title: A Distributed Data-Parallel PyTorch Implementation of the Distributed Shampoo Optimizer for Training Neural Networks At-Scale
• Authors: Hao-Jun Michael Shi, Tsung-Hsien Lee, Shintaro Iwasaki, Jose Gallego-Posada, Zhijing Li, Kaushik Rangadurai, Dheevatsa Mudigere, and Michael Rabbat
• Abstract summary: Shampoo is an online and optimization algorithm belonging to the AdaGrad family of methods for training neural networks. We provide a complete description of the algorithm as well as the performance optimizations that our implementation leverages to train deep networks at-scale in PyTorch.
• Score: 5.206015354543744
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Shampoo is an online and stochastic optimization algorithm belonging to the AdaGrad family of methods for training neural networks. It constructs a block-diagonal preconditioner where each block consists of a coarse Kronecker product approximation to full-matrix AdaGrad for each parameter of the neural network. In this work, we provide a complete description of the algorithm as well as the performance optimizations that our implementation leverages to train deep networks at-scale in PyTorch. Our implementation enables fast multi-GPU distributed data-parallel training by distributing the memory and computation associated with blocks of each parameter via PyTorch's DTensor data structure and performing an AllGather primitive on the computed search directions at each iteration. This major performance enhancement enables us to achieve at most a 10% performance reduction in per-step wall-clock time compared against standard diagonal-scaling-based adaptive gradient methods. We validate our implementation by performing an ablation study on training ImageNet ResNet50, demonstrating Shampoo's superiority over standard training recipes with minimal hyperparameter tuning.

Related papers

• Automated Sizing and Training of Efficient Deep Autoencoders using Second Order Algorithms [0.46040036610482665]
  We propose a multi-step training method for generalized linear classifiers. validation error is minimized by pruning of unnecessary inputs. desired outputs are improved via a method similar to the Ho-Kashyap rule.
  arXiv  Detail & Related papers  (2023-08-11T16:48:31Z)
• Efficient Dataset Distillation Using Random Feature Approximation [109.07737733329019]
  We propose a novel algorithm that uses a random feature approximation (RFA) of the Neural Network Gaussian Process (NNGP) kernel. Our algorithm provides at least a 100-fold speedup over KIP and can run on a single GPU. Our new method, termed an RFA Distillation (RFAD), performs competitively with KIP and other dataset condensation algorithms in accuracy over a range of large-scale datasets.
  arXiv  Detail & Related papers  (2022-10-21T15:56:13Z)
• Neural Nets with a Newton Conjugate Gradient Method on Multiple GPUs [0.0]
  Training deep neural networks consumes increasing computational resource shares in many compute centers. We introduce a novel second-order optimization method that requires the effect of the Hessian on a vector only. We compare the proposed second-order method with two state-of-the-arts on five representative neural network problems.
  arXiv  Detail & Related papers  (2022-08-03T12:38:23Z)
• Online Convolutional Re-parameterization [51.97831675242173]
  We present online convolutional re- parameterization (OREPA), a two-stage pipeline, aiming to reduce the huge training overhead by squeezing the complex training-time block into a single convolution. Compared with the state-of-the-art re-param models, OREPA is able to save the training-time memory cost by about 70% and accelerate the training speed by around 2x. We also conduct experiments on object detection and semantic segmentation and show consistent improvements on the downstream tasks.
  arXiv  Detail & Related papers  (2022-04-02T09:50:19Z)
• A Stochastic Bundle Method for Interpolating Networks [18.313879914379008]
  We propose a novel method for training deep neural networks that are capable of driving the empirical loss to zero. At each iteration our method constructs a maximum linear approximation, known as the bundle of the objective learning approximation.
  arXiv  Detail & Related papers  (2022-01-29T23:02:30Z)
• Instant Neural Graphics Primitives with a Multiresolution Hash Encoding [67.33850633281803]
  We present a versatile new input encoding that permits the use of a smaller network without sacrificing quality. A small neural network is augmented by a multiresolution hash table of trainable feature vectors whose values are optimized through a gradient descent. We achieve a combined speed of several orders of magnitude, enabling training of high-quality neural graphics primitives in a matter of seconds.
  arXiv  Detail & Related papers  (2022-01-16T07:22:47Z)
• Joint inference and input optimization in equilibrium networks [68.63726855991052]
  deep equilibrium model is a class of models that foregoes traditional network depth and instead computes the output of a network by finding the fixed point of a single nonlinear layer. We show that there is a natural synergy between these two settings. We demonstrate this strategy on various tasks such as training generative models while optimizing over latent codes, training models for inverse problems like denoising and inpainting, adversarial training and gradient based meta-learning.
  arXiv  Detail & Related papers  (2021-11-25T19:59:33Z)
• Scheduling Optimization Techniques for Neural Network Training [3.1617796705744547]
  This paper proposes out-of-order (ooo) backprop, an effective scheduling technique for neural network training. We show that the GPU utilization in single-GPU, data-parallel, and pipeline-parallel training can be commonly improve by applying ooo backprop.
  arXiv  Detail & Related papers  (2021-10-03T05:45:06Z)
• Unfolding Projection-free SDP Relaxation of Binary Graph Classifier via GDPA Linearization [59.87663954467815]
  Algorithm unfolding creates an interpretable and parsimonious neural network architecture by implementing each iteration of a model-based algorithm as a neural layer. In this paper, leveraging a recent linear algebraic theorem called Gershgorin disc perfect alignment (GDPA), we unroll a projection-free algorithm for semi-definite programming relaxation (SDR) of a binary graph. Experimental results show that our unrolled network outperformed pure model-based graph classifiers, and achieved comparable performance to pure data-driven networks but using far fewer parameters.
  arXiv  Detail & Related papers  (2021-09-10T07:01:15Z)
• Random Features for the Neural Tangent Kernel [57.132634274795066]
  We propose an efficient feature map construction of the Neural Tangent Kernel (NTK) of fully-connected ReLU network. We show that dimension of the resulting features is much smaller than other baseline feature map constructions to achieve comparable error bounds both in theory and practice.
  arXiv  Detail & Related papers  (2021-04-03T09:08:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.