Direct Quantized Training of Language Models with Stochastic Rounding

• URL: http://arxiv.org/abs/2412.04787v1
• Date: Fri, 06 Dec 2024 05:41:11 GMT
• Title: Direct Quantized Training of Language Models with Stochastic Rounding
• Authors: Kaiyan Zhao, Tsuguchika Tabaru, Kenichi Kobayashi, Takumi Honda, Masafumi Yamazaki, Yoshimasa Tsuruoka,
• Abstract summary: This paper explores the potential of directly updating the quantized low-precision weight matrices without relying on the straight-through estimator during backpropagation. Experimental results on our LLaMA-structured models indicate that training with only low-precision weights is feasible even when they are constrained to ternary values. Our models can also perform inference using ternary weights, showcasing their flexibility in deployment.
• Score: 12.028887152979046
• License:
• Abstract: Although recent quantized Large Language Models (LLMs), such as BitNet, have paved the way for significant reduction in memory usage during deployment with binary or ternary weights, training these models still demands substantial memory footprints. This is partly because high-precision (i.e., unquantized) weight matrices required for straight-through estimation must be maintained throughout the whole training process. To address this, we explore the potential of directly updating the quantized low-precision weight matrices without relying on the straight-through estimator during backpropagation, thereby saving memory usage during training. Specifically, we employ a stochastic rounding technique to minimize information loss caused by the use of low-bit weights throughout training. Experimental results on our LLaMA-structured models indicate that (1) training with only low-precision weights is feasible even when they are constrained to ternary values, (2) extending the bit width to 8 bits results in only a 5% loss degradation compared to BitNet b1.58 while offering the potential for reduced memory usage during training, and (3) our models can also perform inference using ternary weights, showcasing their flexibility in deployment.

Related papers

• The Journey Matters: Average Parameter Count over Pre-training Unifies Sparse and Dense Scaling Laws [51.608402959163925]
  We present the first systematic exploration of optimal sparse pre-training configurations for large language models. We find that initiating pruning at 25% of total training compute and concluding at 75% achieves near-optimal final evaluation loss. We propose a new scaling law that modifies the Chinchilla scaling law to use the average parameter count over pre-training.
  arXiv  Detail & Related papers  (2025-01-21T20:23:22Z)
• Scaling Laws for Precision [73.24325358259753]
  We devise "precision-aware" scaling laws for both training and inference. For inference, we find that the degradation introduced by post-training quantization increases as models are trained on more data. For training, our scaling laws allow us to predict the loss of a model with different parts in different precisions.
  arXiv  Detail & Related papers  (2024-11-07T00:10:10Z)
• Reusing Pretrained Models by Multi-linear Operators for Efficient Training [65.64075958382034]
  Training large models from scratch usually costs a substantial amount of resources. Recent studies such as bert2BERT and LiGO have reused small pretrained models to initialize a large model. We propose a method that linearly correlates each weight of the target model to all the weights of the pretrained model.
  arXiv  Detail & Related papers  (2023-10-16T06:16:47Z)
• Winner-Take-All Column Row Sampling for Memory Efficient Adaptation of Language Model [89.8764435351222]
  We propose a new family of unbiased estimators called WTA-CRS, for matrix production with reduced variance. Our work provides both theoretical and experimental evidence that, in the context of tuning transformers, our proposed estimators exhibit lower variance compared to existing ones.
  arXiv  Detail & Related papers  (2023-05-24T15:52:08Z)
• SPDF: Sparse Pre-training and Dense Fine-tuning for Large Language Models [4.114555639014612]
  We show the benefits of using unstructured weight sparsity to train only a subset of weights during pre-training. We demonstrate that we can induce up to 75% sparsity into a 1.3B parameter GPT-3 XL model resulting in a 2.5x reduction in pre-training FLOPs.
  arXiv  Detail & Related papers  (2023-03-18T17:56:01Z)
• Quantized Neural Networks for Low-Precision Accumulation with Guaranteed Overflow Avoidance [68.8204255655161]
  We introduce a quantization-aware training algorithm that guarantees avoiding numerical overflow when reducing the precision of accumulators during inference. We evaluate our algorithm across multiple quantized models that we train for different tasks, showing that our approach can reduce the precision of accumulators while maintaining model accuracy with respect to a floating-point baseline.
  arXiv  Detail & Related papers  (2023-01-31T02:46:57Z)
• Enabling Binary Neural Network Training on the Edge [7.32770338248516]
  Existing binary neural network training methods require concurrent storage of high-precision activations for all layers. We introduce a low-cost binary neural network training strategy exhibiting sizable memory footprint reductions. We also demonstrate from-scratch ImageNet training of binarized ResNet-18, achieving a 3.78$times$ memory reduction.
  arXiv  Detail & Related papers  (2021-02-08T15:06:41Z)
• Predicting Training Time Without Training [120.92623395389255]
  We tackle the problem of predicting the number of optimization steps that a pre-trained deep network needs to converge to a given value of the loss function. We leverage the fact that the training dynamics of a deep network during fine-tuning are well approximated by those of a linearized model. We are able to predict the time it takes to fine-tune a model to a given loss without having to perform any training.
  arXiv  Detail & Related papers  (2020-08-28T04:29:54Z)
• Improving compute efficacy frontiers with SliceOut [31.864949424541344]
  We introduce SliceOut -- a dropout-inspired scheme to train deep learning models faster without impacting final test accuracy. At test time, turning off SliceOut performs an implicit ensembling across a linear number of architectures that preserves test accuracy. This leads to faster processing of large computational workloads overall, and significantly reduce the resulting energy consumption and CO2emissions.
  arXiv  Detail & Related papers  (2020-07-21T15:59:09Z)

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.