English
Related papers

Related papers: Effects of Parameter Norm Growth During Transforme…

200 papers

Over-parameterized models, such as large deep networks, often exhibit a double descent phenomenon, whereas a function of model size, error first decreases, increases, and decreases at last. This intriguing double descent behavior also…

Machine Learning · Computer Science 2020-09-22 Reinhard Heckel , Fatih Furkan Yilmaz

Self-attention is key to the remarkable success of transformers in sequence modeling tasks including many applications in natural language processing and computer vision. Like neural network layers, these attention mechanisms are often…

Machine Learning · Computer Science 2024-06-21 Tan M. Nguyen , Tam Nguyen , Nhat Ho , Andrea L. Bertozzi , Richard G. Baraniuk , Stanley J. Osher

While the Transformer architecture has achieved remarkable success across various domains, a thorough theoretical foundation explaining its optimization dynamics is yet to be fully developed. In this study, we aim to bridge this…

Machine Learning · Computer Science 2024-11-13 Bingqing Song , Boran Han , Shuai Zhang , Jie Ding , Mingyi Hong

Recent progress in NLP witnessed the development of large-scale pre-trained language models (GPT, BERT, XLNet, etc.) based on Transformer (Vaswani et al. 2017), and in a range of end tasks, such models have achieved state-of-the-art…

Computation and Language · Computer Science 2019-11-12 Pengxiang Cheng , Katrin Erk

When training neural networks as classifiers, it is common to observe an increase in average test loss while still maintaining or improving the overall classification accuracy on the same dataset. In spite of the ubiquity of this…

Machine Learning · Computer Science 2021-03-16 Arthur E. W. Venter , Marthinus W. Theunissen , Marelie H. Davel

A candidate explanation of the good empirical performance of deep neural networks is the implicit regularization effect of first order optimization methods. Inspired by this, we prove a convergence theorem for nonconvex composite…

Machine Learning · Computer Science 2023-02-14 Dávid Terjék , Diego González-Sánchez

The loss surface of an overparameterized neural network (NN) possesses many global minima of zero training error. We explain how common variants of the standard NN training procedure change the minimizer obtained. First, we make explicit…

Machine Learning · Statistics 2023-09-26 Manuel Nonnenmacher , David Reeb , Ingo Steinwart

In the context of over-parameterization, there is a line of work demonstrating that randomly initialized (stochastic) gradient descent (GD) converges to a globally optimal solution at a linear convergence rate for the quadratic loss…

Machine Learning · Computer Science 2025-06-16 Xianliang Xu , Ting Du , Wang Kong , Bin Shan , Ye Li , Zhongyi Huang

Recent years have seen a proliferation of attention mechanisms and the rise of Transformers in Natural Language Generation (NLG). Previously, state-of-the-art NLG architectures such as RNN and LSTM ran into vanishing gradient problems; as…

Computation and Language · Computer Science 2021-02-17 M. Onat Topal , Anil Bas , Imke van Heerden

Deep neural networks (DNNs) are quantized for efficient inference on resource-constrained platforms. However, training deep learning models with low-precision weights and activations involves a demanding optimization task, which calls for…

Machine Learning · Computer Science 2021-05-25 Ziang Long , Penghang Yin , Jack Xin

Large language models (LLMs) can learn from a few demonstrations provided at inference time. We study this in-context learning phenomenon through the lens of Gaussian Processes (GPs). We build controlled experiments where models observe…

Machine Learning · Computer Science 2026-02-13 Elif Akata , Konstantinos Voudouris , Vincent Fortuin , Eric Schulz

When fine-tuning Deep Neural Networks (DNNs) to new data, DNNs are prone to overwriting network parameters required for task-specific functionality on previously learned tasks, resulting in a loss of performance on those tasks. We propose…

Machine Learning · Computer Science 2025-01-22 Christopher Angelini , Nidhal Bouaynaya

The weight initialization and the activation function of deep neural networks have a crucial impact on the performance of the training procedure. An inappropriate selection can lead to the loss of information of the input during forward…

Machine Learning · Statistics 2019-05-28 Soufiane Hayou , Arnaud Doucet , Judith Rousseau

There is an ongoing debate on whether neural networks can grasp the quasi-regularities in languages like humans. In a typical quasi-regularity task, English past tense inflections, the neural network model has long been criticized that it…

Computation and Language · Computer Science 2023-05-16 Xiaomeng Ma , Lingyu Gao

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

We consider the dynamics of gradient descent (GD) in overparameterized single hidden layer neural networks with a squared loss function. Recently, it has been shown that, under some conditions, the parameter values obtained using GD achieve…

Machine Learning · Computer Science 2021-05-17 Siddhartha Satpathi , R Srikant

Modern neural networks are usually highly over-parameterized. Behind the wide usage of over-parameterized networks is the belief that, if the data are simple, then the trained network will be automatically equivalent to a simple predictor.…

Machine Learning · Statistics 2025-04-14 Chenyang Zhang , Peifeng Gao , Difan Zou , Yuan Cao

Machine learning requires exuberant amounts of data and computation. Also, models require equally excessive growth in the number of parameters. It is, therefore, sensible to look for technologies that reduce these demands on resources.…

Machine Learning · Computer Science 2023-03-29 Danko Nikolić , Davor Andrić , Vjekoslav Nikolić

A recent line of research has shown that gradient-based algorithms with random initialization can converge to the global minima of the training loss for over-parameterized (i.e., sufficiently wide) deep neural networks. However, the…

Machine Learning · Computer Science 2019-06-12 Difan Zou , Quanquan Gu

Unsupervised pretraining and dropout have been well studied, especially with respect to regularization and output consistency. However, our understanding about the explicit convergence rates of the parameter estimates, and their dependence…

Machine Learning · Computer Science 2017-02-23 Vamsi K. Ithapu , Sathya Ravi , Vikas Singh