English
Related papers

Related papers: Stability of Transformers under Layer Normalizatio…

200 papers

From the perspective of the layer normalization (LN) positions, the architectures of Transformers can be categorized into two types: Post-LN and Pre-LN. Recent Transformers tend to be Pre-LN because, in Post-LN with deep Transformers (e.g.,…

Machine Learning · Computer Science 2023-05-29 Sho Takase , Shun Kiyono , Sosuke Kobayashi , Jun Suzuki

The Transformer is widely used in natural language processing tasks. To train a Transformer however, one usually needs a carefully designed learning rate warm-up stage, which is shown to be crucial to the final performance but will slow…

Machine Learning · Computer Science 2020-06-30 Ruibin Xiong , Yunchang Yang , Di He , Kai Zheng , Shuxin Zheng , Chen Xing , Huishuai Zhang , Yanyan Lan , Liwei Wang , Tie-Yan Liu

Transformers have proved effective in many NLP tasks. However, their training requires non-trivial efforts regarding designing cutting-edge optimizers and learning rate schedulers carefully (e.g., conventional SGD fails to train…

Machine Learning · Computer Science 2023-10-03 Liyuan Liu , Xiaodong Liu , Jianfeng Gao , Weizhu Chen , Jiawei Han

Despite their impressive performance, contemporary neural networks often lack structural safeguards that promote stable learning and interpretable behavior. In this work, we introduce a reformulation of layer-level transformations that…

Machine Learning · Computer Science 2025-08-04 Saleh Nikooroo , Thomas Engel

In spite of their huge success, transformer models remain difficult to scale in depth. In this work, we develop a unified signal propagation theory and provide formulae that govern the moments of the forward and backward signal through the…

Computation and Language · Computer Science 2024-07-19 Akhil Kedia , Mohd Abbas Zaidi , Sushil Khyalia , Jungho Jung , Harshith Goka , Haejun Lee

We introduce a new technique for gradient normalization during neural network training. The gradients are rescaled during the backward pass using normalization layers introduced at certain points within the network architecture. These…

Machine Learning · Computer Science 2021-06-18 Alejandro Cabana , Luis F. Lago-Fernández

In this paper, we propose a simple yet effective method to stabilize extremely deep Transformers. Specifically, we introduce a new normalization function (DeepNorm) to modify the residual connection in Transformer, accompanying with…

Computation and Language · Computer Science 2022-03-02 Hongyu Wang , Shuming Ma , Li Dong , Shaohan Huang , Dongdong Zhang , Furu Wei

The success of Large Language Models (LLMs) hinges on the stable training of deep Transformer architectures. A critical design choice is the placement of normalization layers, leading to a fundamental trade-off: the ``PreNorm'' architecture…

Computation and Language · Computer Science 2026-02-02 Chao Wang , Bei Li , Jiaqi Zhang , Xinyu Liu , Yuchun Fan , Linkun Lyu , Xin Chen , Jingang Wang , Tong Xiao , Peng Pei , Xunliang Cai

Scaling deep reinforcement learning networks is challenging and often results in degraded performance, yet the root causes of this failure mode remain poorly understood. Several recent works have proposed mechanisms to address this, but…

Teams that have trained large Transformer-based models have reported training instabilities at large scale that did not appear when training with the same hyperparameters at smaller scales. Although the causes of such instabilities are of…

Traditional analyses of gradient descent optimization show that, when the largest eigenvalue of the loss Hessian - often referred to as the sharpness - is below a critical learning-rate threshold, then training is 'stable' and training loss…

Machine Learning · Computer Science 2024-12-24 Lawrence Wang , Stephen J. Roberts

Despite their nearly universal adoption for large language models, the internal workings of transformers are not well understood. We aim to better understand the impact of removing or reorganizing information throughout the layers of a…

Computation and Language · Computer Science 2025-02-14 Qi Sun , Marc Pickett , Aakash Kumar Nain , Llion Jones

Gradient dynamics play a central role in determining the stability and generalization of deep neural networks. In this work, we provide an empirical analysis of how variance and standard deviation of gradients evolve during training,…

Machine Learning · Computer Science 2025-09-09 Vincent-Daniel Yun

A recent line of work has established intriguing connections between the generalization/compression properties of a deep neural network (DNN) model and the so-called layer weights' stable ranks. Intuitively, the latter are indicators of the…

Machine Learning · Computer Science 2021-10-07 Bogdan Georgiev , Lukas Franken , Mayukh Mukherjee , Georgios Arvanitidis

In this paper we address the issue of output instability of deep neural networks: small perturbations in the visual input can significantly distort the feature embeddings and output of a neural network. Such instability affects many deep…

Computer Vision and Pattern Recognition · Computer Science 2016-04-18 Stephan Zheng , Yang Song , Thomas Leung , Ian Goodfellow

Transformer models have emerged as fundamental tools across various scientific and engineering disciplines, owing to their outstanding performance in diverse applications. Despite this empirical success, the theoretical foundations of…

Machine Learning · Computer Science 2026-04-14 Zhen Qin , Jinxin Zhou , Jiachen Jiang , Zhihui Zhu

Existing analyses of neural network training often operate under the unrealistic assumption of an extremely small learning rate. This lies in stark contrast to practical wisdom and empirical studies, such as the work of J. Cohen et al.…

Machine Learning · Computer Science 2023-10-20 Kwangjun Ahn , Sébastien Bubeck , Sinho Chewi , Yin Tat Lee , Felipe Suarez , Yi Zhang

Transformers trained in low precision can suffer forward-error amplification. We give a first-order, module-wise theory that predicts when and where errors grow. For self-attention we derive a per-layer bound that factorizes into three…

Machine Learning · Computer Science 2025-10-28 Jinwoo Baek

Normalization layers were introduced to stabilize and accelerate training, yet their influence is critical already at initialization, where they shape signal propagation and output statistics before parameters adapt to data. In practice,…

Machine Learning · Computer Science 2026-04-03 Emanuele Francazi , Francesco Pinto , Aurelien Lucchi , Marco Baity-Jesi

We examine the stability of loss-minimizing training processes that are used for deep neural networks (DNN) and other classifiers. While a classifier is optimized during training through a so-called loss function, the performance of…

Analysis of PDEs · Mathematics 2020-10-05 Leonid Berlyand , Pierre-Emmanuel Jabin , C. Alex Safsten
‹ Prev 1 2 3 10 Next ›