English
Related papers

Related papers: Stability of Transformers under Layer Normalizatio…

200 papers

Graph neural networks are experiencing a surge of popularity within the machine learning community due to their ability to adapt to non-Euclidean domains and instil inductive biases. Despite this, their stability, i.e., their robustness to…

Machine Learning · Computer Science 2021-02-19 Henry Kenlay , Dorina Thanou , Xiaowen Dong

Background: Recent developments have made it possible to accelerate neural networks training significantly using large batch sizes and data parallelism. Training in an asynchronous fashion, where delay occurs, can make training even more…

Machine Learning · Computer Science 2020-02-14 Niv Giladi , Mor Shpigel Nacson , Elad Hoffer , Daniel Soudry

Of all the vector fields surrounding the minima of recurrent learning setups, the gradient field with its exploding and vanishing updates appears a poor choice for optimization, offering little beyond efficient computability. We seek to…

Machine Learning · Computer Science 2024-05-06 Patrick Schnell , Nils Thuerey

This study investigates the layerwise importance of feed-forward networks (FFNs) in Transformer-based language models during pretraining. We introduce an experimental approach that, while maintaining the total parameter count, increases the…

Computation and Language · Computer Science 2025-08-26 Wataru Ikeda , Kazuki Yano , Ryosuke Takahashi , Jaesung Lee , Keigo Shibata , Jun Suzuki

Generalization analyses of deep learning typically assume that the training converges to a fixed point. But, recent results indicate that in practice, the weights of deep neural networks optimized with stochastic gradient descent often…

Machine Learning · Computer Science 2022-08-22 Nisha Chandramoorthy , Andreas Loukas , Khashayar Gatmiry , Stefanie Jegelka

While training error of most deep neural networks degrades as the depth of the network increases, residual networks appear to be an exception. We show that the main reason for this is the Lyapunov stability of the gradient descent…

Machine Learning · Computer Science 2018-03-23 Kamil Nar , Shankar Sastry

Fine-tuning over large pretrained language models (PLMs) has established many state-of-the-art results. Despite its superior performance, such fine-tuning can be unstable, resulting in significant variance in performance and potential risks…

Computation and Language · Computer Science 2022-10-20 Chenghao Yang , Xuezhe Ma

Transformer networks have seen great success in natural language processing and machine vision, where task objectives such as next word prediction and image classification benefit from nuanced context sensitivity across high-dimensional…

Machine Learning · Computer Science 2022-12-13 Yuxuan Li , James L. McClelland

Understanding the implicit bias of training algorithms is of crucial importance in order to explain the success of overparametrised neural networks. In this paper, we study the dynamics of stochastic gradient descent over diagonal linear…

Machine Learning · Computer Science 2021-12-08 Scott Pesme , Loucas Pillaud-Vivien , Nicolas Flammarion

The overall performance or expected excess risk of an iterative machine learning algorithm can be decomposed into training error and generalization error. While the former is controlled by its convergence analysis, the latter can be tightly…

Machine Learning · Statistics 2018-04-06 Yuansi Chen , Chi Jin , Bin Yu

There is a growing interest in developing data-driven subgrid-scale (SGS) models for large-eddy simulation (LES) using machine learning (ML). In a priori (offline) tests, some recent studies have found ML-based data-driven SGS models that…

Fluid Dynamics · Physics 2022-03-23 Yifei Guan , Ashesh Chattopadhyay , Adam Subel , Pedram Hassanzadeh

Deep ResNets are recognized for achieving state-of-the-art results in complex machine learning tasks. However, the remarkable performance of these architectures relies on a training procedure that needs to be carefully crafted to avoid…

Machine Learning · Computer Science 2025-03-04 Pierre Marion , Adeline Fermanian , Gérard Biau , Jean-Philippe Vert

As an emerging type of Neural Networks (NNs), Transformers are used in many domains ranging from Natural Language Processing to Autonomous Driving. In this paper, we study the robustness problem of Transformers, a key characteristic as low…

Machine Learning · Computer Science 2024-12-03 Brian Hsuan-Cheng Liao , Chih-Hong Cheng , Hasan Esen , Alois Knoll

An influential line of recent work has focused on the generalization properties of unregularized gradient-based learning procedures applied to separable linear classification with exponentially-tailed loss functions. The ability of such…

Machine Learning · Computer Science 2022-06-24 Matan Schliserman , Tomer Koren

Stacking is a general approach for combining multiple models toward greater predictive accuracy. It has found various application across different domains, ensuing from its meta-learning nature. Our understanding, nevertheless, on how and…

Machine Learning · Computer Science 2019-01-29 Nino Arsov , Martin Pavlovski , Ljupco Kocarev

The internal representations learned by deep networks are often sensitive to architecture-specific choices, raising questions about the stability, alignment, and transferability of learned structure across models. In this paper, we…

Machine Learning · Computer Science 2025-08-06 Saleh Nikooroo , Thomas Engel

Transformers have achieved extraordinary success in modern machine learning due to their excellent ability to handle sequential data, especially in next-token prediction (NTP) tasks. However, the theoretical understanding of their…

Machine Learning · Computer Science 2024-10-01 Ruiquan Huang , Yingbin Liang , Jing Yang

Looped transformers promise test-time compute scaling by spending more iterations on harder problems, but it remains unclear which architectural choices let them extrapolate to harder problems at test time rather than memorize…

Machine Learning · Computer Science 2026-04-23 Asher Labovich

We investigate forward signal propagation and gradient back propagation in deep, randomly initialized transformers, yielding simple necessary and sufficient conditions on initialization hyperparameters that ensure trainability of deep…

Disordered Systems and Neural Networks · Physics 2024-03-06 Aditya Cowsik , Tamra Nebabu , Xiao-Liang Qi , Surya Ganguli

Adversarially robust models are locally smooth around each data sample so that small perturbations cannot drastically change model outputs. In modern systems, such smoothness is usually obtained via Adversarial Training, which explicitly…

Machine Learning · Computer Science 2024-10-01 Adrián Rodríguez-Muñoz , Tongzhou Wang , Antonio Torralba