English
Related papers

Related papers: Correlations Are Ruining Your Gradient Descent

200 papers

We analyze speed of convergence to global optimum for gradient descent training a deep linear neural network (parameterized as $x \mapsto W_N W_{N-1} \cdots W_1 x$) by minimizing the $\ell_2$ loss over whitened data. Convergence at a linear…

Machine Learning · Computer Science 2019-10-29 Sanjeev Arora , Nadav Cohen , Noah Golowich , Wei Hu

We propose a new technique that boosts the convergence of training generative adversarial networks. Generally, the rate of training deep models reduces severely after multiple iterations. A key reason for this phenomenon is that a deep…

Machine Learning · Statistics 2018-06-15 Atsushi Nitanda , Taiji Suzuki

We study the properties of alignment, a form of implicit regularization, in linear neural networks under gradient descent. We define alignment for fully connected networks with multidimensional outputs and show that it is a natural…

Machine Learning · Computer Science 2020-06-18 Adityanarayanan Radhakrishnan , Eshaan Nichani , Daniel Bernstein , Caroline Uhler

In this paper, we study the implicit regularization of the gradient descent algorithm in homogeneous neural networks, including fully-connected and convolutional neural networks with ReLU or LeakyReLU activations. In particular, we study…

Machine Learning · Computer Science 2021-01-01 Kaifeng Lyu , Jian Li

Normalization layers are widely used in deep neural networks to stabilize training. In this paper, we consider the training of convolutional neural networks with gradient descent on a single training example. This optimization problem…

Machine Learning · Computer Science 2019-07-24 Zhenwei Dai , Reinhard Heckel

Recent years have seen a flurry of activities in designing provably efficient nonconvex procedures for solving statistical estimation problems. Due to the highly nonconvex nature of the empirical loss, state-of-the-art procedures often…

Machine Learning · Computer Science 2020-06-09 Cong Ma , Kaizheng Wang , Yuejie Chi , Yuxin Chen

Relieving the reliance of neural network training on a global back-propagation (BP) has emerged as a notable research topic due to the biological implausibility and huge memory consumption caused by BP. Among the existing solutions, local…

Machine Learning · Computer Science 2024-06-11 Yibo Yang , Xiaojie Li , Motasem Alfarra , Hasan Hammoud , Adel Bibi , Philip Torr , Bernard Ghanem

Recently, several studies have proven the global convergence and generalization abilities of the gradient descent method for two-layer ReLU networks. Most studies especially focused on the regression problems with the squared loss function,…

Machine Learning · Statistics 2020-03-19 Atsushi Nitanda , Geoffrey Chinot , Taiji Suzuki

The backpropagation algorithm has long been the canonical training method for neural networks. Modern paradigms are implicitly optimized for it, and numerous guidelines exist to ensure its proper use. Recently, synthetic gradients methods…

Machine Learning · Statistics 2019-06-12 Julien Launay , Iacopo Poli , Florent Krzakala

Training a neural network using backpropagation algorithm requires passing error gradients sequentially through the network. The backward locking prevents us from updating network layers in parallel and fully leveraging the computing…

Machine Learning · Computer Science 2019-05-30 Zhouyuan Huo , Bin Gu , Heng Huang

Natural gradient descent (NGD) provided deep insights and powerful tools to deep neural networks. However the computation of Fisher information matrix becomes more and more difficult as the network structure turns large and complex. This…

Machine Learning · Computer Science 2021-09-22 Weihua Liu , Xiabi Liu

Neural networks are trained primarily based on their inputs and outputs, without regard for their internal mechanisms. These neglected mechanisms determine properties that are critical for safety, like (i) transparency; (ii) the absence of…

Machine Learning · Computer Science 2024-12-02 Alex Cloud , Jacob Goldman-Wetzler , Evžen Wybitul , Joseph Miller , Alexander Matt Turner

One of the mysteries in the success of neural networks is randomly initialized first order methods like gradient descent can achieve zero training loss even though the objective function is non-convex and non-smooth. This paper demystifies…

Machine Learning · Computer Science 2019-02-06 Simon S. Du , Xiyu Zhai , Barnabas Poczos , Aarti Singh

Backpropagation is the core learning mechanism underlying deep learning. However, whether and how this algorithm is implemented in the brain remains highly debated. In particular, while forward activations of pretrained models reliably map…

Even for the gradient descent (GD) method applied to neural network training, understanding its optimization dynamics, including convergence rate, iterate trajectories, function value oscillations, and especially its implicit acceleration,…

Machine Learning · Computer Science 2026-05-22 Alexander Tyurin

Deep learning approaches, known for their ability to model complex relationships and fast execution, are increasingly being applied to solve large optimization problems. However, existing methods often face challenges in simultaneously…

Optimization and Control · Mathematics 2025-12-16 Zisheng Zhou , Dengyu Zheng , Zirui Chen , Shixiang Chen

Stochastic Gradient Descent (SGD) is the main approach to optimizing neural networks. Several generalization properties of deep networks, such as convergence to a flatter minima, are believed to arise from SGD. This article explores the…

Machine Learning · Computer Science 2024-12-05 Aditya Shah , Aditya Challa , Sravan Danda , Archana Mathur , Snehanshu Saha

Gradient descent prevails in artificial neural network training, but seems inept for spiking neural networks as small parameter changes can cause sudden, disruptive (dis-)appearances of spikes. Here, we demonstrate exact gradient descent…

Neurons and Cognition · Quantitative Biology 2025-01-29 Christian Klos , Raoul-Martin Memmesheimer

Any gradient descent optimization requires to choose a learning rate. With deeper and deeper models, tuning that learning rate can easily become tedious and does not necessarily lead to an ideal convergence. We propose a variation of the…

Machine Learning · Statistics 2018-04-10 Mathieu Ravaut , Satya Gorti

We consider the problem of approximating a function by an element of a nonlinear manifold which admits a differentiable parametrization, typical examples being neural networks with differentiable activation functions or tensor networks.…

Machine Learning · Computer Science 2026-04-20 Anthony Nouy , Agustín Somacal