No Parameters Left Behind: Sensitivity Guided Adaptive Learning Rate for Training Large Transformer Models

• URL: http://arxiv.org/abs/2202.02664v1
• Date: Sun, 6 Feb 2022 00:22:28 GMT
• Title: No Parameters Left Behind: Sensitivity Guided Adaptive Learning Rate for Training Large Transformer Models
• Authors: Chen Liang, Haoming Jiang, Simiao Zuo, Pengcheng He, Xiaodong Liu, Jianfeng Gao, Weizhu Chen, Tuo Zhao
• Abstract summary: We propose a novel training strategy that encourages all parameters to be trained sufficiently. A parameter with low sensitivity is redundant, and we improve its fitting by increasing its learning rate. In contrast, a parameter with high sensitivity is well-trained and we regularize it by decreasing its learning rate to prevent further overfitting.
• Score: 132.90062129639705
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent research has shown the existence of significant redundancy in large Transformer models. One can prune the redundant parameters without significantly sacrificing the generalization performance. However, we question whether the redundant parameters could have contributed more if they were properly trained. To answer this question, we propose a novel training strategy that encourages all parameters to be trained sufficiently. Specifically, we adaptively adjust the learning rate for each parameter according to its sensitivity, a robust gradient-based measure reflecting this parameter's contribution to the model performance. A parameter with low sensitivity is redundant, and we improve its fitting by increasing its learning rate. In contrast, a parameter with high sensitivity is well-trained, and we regularize it by decreasing its learning rate to prevent further overfitting. We conduct extensive experiments on natural language understanding, neural machine translation, and image classification to demonstrate the effectiveness of the proposed schedule. Analysis shows that the proposed schedule indeed reduces the redundancy and improves generalization performance.

Related papers

• Scaling Exponents Across Parameterizations and Optimizers [94.54718325264218]
  We propose a new perspective on parameterization by investigating a key assumption in prior work. Our empirical investigation includes tens of thousands of models trained with all combinations of threes. We find that the best learning rate scaling prescription would often have been excluded by the assumptions in prior work.
  arXiv  Detail & Related papers  (2024-07-08T12:32:51Z)
• Sine Activated Low-Rank Matrices for Parameter Efficient Learning [25.12262017296922]
  We propose a novel theoretical framework that integrates a sinusoidal function within the low-rank decomposition process. Our method proves to be an enhancement for existing low-rank models, as evidenced by its successful application in Vision Transformers (ViT), Large Language Models (LLMs), Neural Radiance Fields (NeRF)
  arXiv  Detail & Related papers  (2024-03-28T08:58:20Z)
• Low-Rank Rescaled Vision Transformer Fine-Tuning: A Residual Design Approach [17.678759882763078]
  Fine-tuning for pre-trained Vision Transformers aims to adeptly tailor a model to downstream tasks. Striking a balance between retaining the generalizable representation capacity of the pre-trained model and acquiring task-specific features is a key challenge. We propose a Residual-based Low-Rank Rescaling (RLRR) fine-tuning strategy.
  arXiv  Detail & Related papers  (2024-03-28T00:14:53Z)
• Re-parameterized Low-rank Prompt: Generalize a Vision-Language Model within 0.5K Parameters [75.28536311904489]
  We develop a new type of prompt, Re- parameterized Low-rank Prompt (RLP), for both efficient and effective adaptation. On a series of tasks over 11 datasets, RLP significantly increases the average downstream accuracy of classic prompt tuning by up to 5.25% using merely 0.5K parameters.
  arXiv  Detail & Related papers  (2023-12-17T20:42:43Z)
• Sparse Low-rank Adaptation of Pre-trained Language Models [79.74094517030035]
  We introduce sparse low-rank adaptation (SoRA) that enables dynamic adjustments to the intrinsic rank during the adaptation process. Our approach strengthens the representation power of LoRA by initializing it with a higher rank, while efficiently taming a temporarily increased number of parameters. Our experimental results demonstrate that SoRA can outperform other baselines even with 70% retained parameters and 70% training time.
  arXiv  Detail & Related papers  (2023-11-20T11:56:25Z)
• The Importance of Being Parameters: An Intra-Distillation Method for Serious Gains [13.579368172149135]
  We argue that redundant parameters can be trained to make beneficial contributions. We propose a general task-agnostic method, namely intra-distillation, appended to the regular training loss to balance sensitivity. Our experiments show the strong effectiveness of our methods on machine translation, natural language understanding, and zero-shot cross-lingual transfer.
  arXiv  Detail & Related papers  (2022-05-23T16:01:46Z)
• Adaptive Gradient Method with Resilience and Momentum [120.83046824742455]
  We propose an Adaptive Gradient Method with Resilience and Momentum (AdaRem) AdaRem adjusts the parameter-wise learning rate according to whether the direction of one parameter changes in the past is aligned with the direction of the current gradient. Our method outperforms previous adaptive learning rate-based algorithms in terms of the training speed and the test error.
  arXiv  Detail & Related papers  (2020-10-21T14:49:00Z)
• 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)

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.