Optimizing Large Model Training through Overlapped Activation Recomputation
- URL: http://arxiv.org/abs/2406.08756v2
- Date: Thu, 27 Jun 2024 12:45:38 GMT
- Title: Optimizing Large Model Training through Overlapped Activation Recomputation
- Authors: Ping Chen, Wenjie Zhang, Shuibing He, Yingjie Gu, Zhuwei Peng, Kexin Huang, Xuan Zhan, Weijian Chen, Yi Zheng, Zhefeng Wang, Yanlong Yin, Gang Chen,
- Abstract summary: Existing recomputation approaches may incur up to 40% overhead when training real-world models.
This is because they are executed on demand in the critical training path.
We design a new recomputation framework, Lynx, to reduce the overhead by overlapping the recomputation with communication occurring in training pipelines.
- Score: 24.461674158317578
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Large model training has been using recomputation to alleviate the memory pressure and pipelining to exploit the parallelism of data, tensor, and devices. The existing recomputation approaches may incur up to 40% overhead when training real-world models, e.g., the GPT model with 22B parameters. This is because they are executed on demand in the critical training path. In this paper, we design a new recomputation framework, Lynx, to reduce the overhead by overlapping the recomputation with communication occurring in training pipelines. It consists of an optimal scheduling algorithm (OPT) and a heuristic-based scheduling algorithm (HEU). OPT achieves a global optimum but suffers from a long search time. HEU was designed based on our observation that there are identical structures in large DNN models so that we can apply the same scheduling policy to all identical structures. HEU achieves a local optimum but reduces the search time by 99% compared to OPT. Our comprehensive evaluation using GPT models with 1.3B-20B parameters shows that both OPT and HEU outperform the state-of-the-art recomputation approaches (e.g., Megatron-LM and Checkmake) by 1.02-1.53x. HEU achieves a similar performance as OPT with a search time of 0.16s on average.
Related papers
- Stabilizing Subject Transfer in EEG Classification with Divergence
Estimation [17.924276728038304]
We propose several graphical models to describe an EEG classification task.
We identify statistical relationships that should hold true in an idealized training scenario.
We design regularization penalties to enforce these relationships in two stages.
arXiv Detail & Related papers (2023-10-12T23:06:52Z) - Trainable Projected Gradient Method for Robust Fine-tuning [36.470333094917436]
We propose Trainable Projected Gradient Method (TPGM) to automatically learn the constraint imposed for each layer for a fine-grained fine-tuning regularization.
This is motivated by formulating fine-tuning as a bi-level constrained optimization problem.
We show that TPGM outperforms existing fine-tuning methods in OOD performance while matching the best in-distribution (ID) performance.
arXiv Detail & Related papers (2023-03-19T17:30:44Z) - Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language
Models [107.05966685291067]
We propose test-time prompt tuning (TPT) to learn adaptive prompts on the fly with a single test sample.
TPT improves the zero-shot top-1 accuracy of CLIP by 3.6% on average.
In evaluating cross-dataset generalization with unseen categories, TPT performs on par with the state-of-the-art approaches that use additional training data.
arXiv Detail & Related papers (2022-09-15T17:55:11Z) - DPTDR: Deep Prompt Tuning for Dense Passage Retrieval [53.217524851268216]
Deep prompt tuning (DPT) has gained great success in most natural language processing(NLP) tasks.
However, it is not well-investigated in dense retrieval where fine-tuning(FT) still dominates.
We propose two model-agnostic and task-agnostic strategies for DPT-based retrievers, namely retrieval-oriented intermediate pretraining and unified negative mining.
arXiv Detail & Related papers (2022-08-24T12:55:00Z) - DSEE: Dually Sparsity-embedded Efficient Tuning of Pre-trained Language
Models [152.29364079385635]
As pre-trained models grow bigger, the fine-tuning process can be time-consuming and computationally expensive.
We propose a framework for resource- and parameter-efficient fine-tuning by leveraging the sparsity prior in both weight updates and the final model weights.
Our proposed framework, dubbed Dually Sparsity-Embedded Efficient Tuning (DSEE), aims to achieve two key objectives: (i) parameter efficient fine-tuning and (ii) resource-efficient inference.
arXiv Detail & Related papers (2021-10-30T03:29:47Z) - Training Recommender Systems at Scale: Communication-Efficient Model and
Data Parallelism [56.78673028601739]
We propose a compression framework called Dynamic Communication Thresholding (DCT) for communication-efficient hybrid training.
DCT reduces communication by at least $100times$ and $20times$ during DP and MP, respectively.
It improves end-to-end training time for a state-of-the-art industrial recommender model by 37%, without any loss in performance.
arXiv Detail & Related papers (2020-10-18T01:44:42Z) - Extrapolation for Large-batch Training in Deep Learning [72.61259487233214]
We show that a host of variations can be covered in a unified framework that we propose.
We prove the convergence of this novel scheme and rigorously evaluate its empirical performance on ResNet, LSTM, and Transformer.
arXiv Detail & Related papers (2020-06-10T08:22:41Z) - Joint Parameter-and-Bandwidth Allocation for Improving the Efficiency of
Partitioned Edge Learning [73.82875010696849]
Machine learning algorithms are deployed at the network edge for training artificial intelligence (AI) models.
This paper focuses on the novel joint design of parameter (computation load) allocation and bandwidth allocation.
arXiv Detail & Related papers (2020-03-10T05:52:15Z)
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.