A Multi-Level Framework for Accelerating Training Transformer Models

• URL: http://arxiv.org/abs/2404.07999v1
• Date: Sun, 7 Apr 2024 03:04:34 GMT
• Title: A Multi-Level Framework for Accelerating Training Transformer Models
• Authors: Longwei Zou, Han Zhang, Yangdong Deng,
• Abstract summary: Training large-scale deep learning models poses an unprecedented demand for computing power. We propose a multi-level framework for training acceleration based on Coalescing, De-coalescing and Interpolation. We prove that the proposed framework reduces the computational cost by about 20% on training BERT/GPT-Base models and up to 51.6% on training the BERT-Large model.
• Score: 5.268960238774481
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The fast growing capabilities of large-scale deep learning models, such as Bert, GPT and ViT, are revolutionizing the landscape of NLP, CV and many other domains. Training such models, however, poses an unprecedented demand for computing power, which incurs exponentially increasing energy cost and carbon dioxide emissions. It is thus critical to develop efficient training solutions to reduce the training costs. Motivated by a set of key observations of inter- and intra-layer similarities among feature maps and attentions that can be identified from typical training processes, we propose a multi-level framework for training acceleration. Specifically, the framework is based on three basic operators, Coalescing, De-coalescing and Interpolation, which can be orchestrated to build a multi-level training framework. The framework consists of a V-cycle training process, which progressively down- and up-scales the model size and projects the parameters between adjacent levels of models via coalescing and de-coalescing. The key idea is that a smaller model that can be trained for fast convergence and the trained parameters provides high-qualities intermediate solutions for the next level larger network. The interpolation operator is designed to break the symmetry of neurons incurred by de-coalescing for better convergence performance. Our experiments on transformer-based language models (e.g. Bert, GPT) as well as a vision model (e.g. DeiT) prove that the proposed framework reduces the computational cost by about 20% on training BERT/GPT-Base models and up to 51.6% on training the BERT-Large model while preserving the performance.

Related papers

• Transferable Post-training via Inverse Value Learning [83.75002867411263]
  We propose modeling changes at the logits level during post-training using a separate neural network (i.e., the value network) After training this network on a small base model using demonstrations, this network can be seamlessly integrated with other pre-trained models during inference. We demonstrate that the resulting value network has broad transferability across pre-trained models of different parameter sizes.
  arXiv  Detail & Related papers  (2024-10-28T13:48:43Z)
• ATOM: Asynchronous Training of Massive Models for Deep Learning in a Decentralized Environment [7.916080032572087]
  atom is a resilient distributed training framework designed for asynchronous training of vast models in a decentralized setting. atom aims to accommodate a complete LLM on one host (peer) through seamlessly model swapping and concurrently trains multiple copies across various peers to optimize training throughput. Our experiments using different GPT-3 model configurations reveal that, in scenarios with suboptimal network connections, atom can enhance training efficiency up to $20 times$ when juxtaposed with the state-of-the-art decentralized pipeline parallelism approaches.
  arXiv  Detail & Related papers  (2024-03-15T17:43:43Z)
• Majority Kernels: An Approach to Leverage Big Model Dynamics for Efficient Small Model Training [32.154166415680066]
  Methods like distillation, compression or quantization help leverage the highly performant large models to induce smaller performant ones. This paper explores the hypothesis that a single training run can simultaneously train a larger model for performance and derive a smaller model for deployment.
  arXiv  Detail & Related papers  (2024-02-07T17:07:41Z)
• Preparing Lessons for Progressive Training on Language Models [75.88952808979087]
  The rapid progress of Transformers in artificial intelligence has come at the cost of increased resource consumption and greenhouse gas emissions. We propose Apollo, which preptextbfares lessons for extextbfpanding textbfoperations by textbflayer functitextbfonality during training of low layers. Experiments demonstrate that Apollo achieves state-of-the-art acceleration ratios, even rivaling methods using pretrained models.
  arXiv  Detail & Related papers  (2024-01-17T13:04:14Z)
• Federated Learning over Hierarchical Wireless Networks: Training Latency Minimization via Submodel Partitioning [15.311309249848739]
  Hierarchical independent submodel training (HIST) is a new FL methodology that aims to address these issues in hierarchical cloud-edge-client networks. We demonstrate how HIST can be augmented with over-the-air computation (AirComp) to further enhance the efficiency of the model aggregation over the edge cells.
  arXiv  Detail & Related papers  (2023-10-27T04:42:59Z)
• One-stop Training of Multiple Capacity Models [74.87789190840527]
  We propose a novel one-stop training framework to jointly train high-capacity and low-capactiy models. Unlike knowledge distillation, where multiple capacity models are trained from scratch separately, our approach integrates supervisions from different capacity models simultaneously.
  arXiv  Detail & Related papers  (2023-05-23T13:44:09Z)
• Optimization-Derived Learning with Essential Convergence Analysis of Training and Hyper-training [52.39882976848064]
  We design a Generalized Krasnoselskii-Mann (GKM) scheme based on fixed-point iterations as our fundamental ODL module. Under the GKM scheme, a Bilevel Meta Optimization (BMO) algorithmic framework is constructed to solve the optimal training and hyper-training variables together.
  arXiv  Detail & Related papers  (2022-06-16T01:50:25Z)
• 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)
• Simultaneous Training of Partially Masked Neural Networks [67.19481956584465]
  We show that it is possible to train neural networks in such a way that a predefined 'core' subnetwork can be split-off from the trained full network with remarkable good performance. We show that training a Transformer with a low-rank core gives a low-rank model with superior performance than when training the low-rank model alone.
  arXiv  Detail & Related papers  (2021-06-16T15:57:51Z)

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.