English
Related papers

Related papers: Revisiting Landscape Analysis in Deep Neural Netwo…

200 papers

A main puzzle of deep neural networks (DNNs) revolves around the apparent absence of "overfitting", defined in this paper as follows: the expected error does not get worse when increasing the number of neurons or of iterations of gradient…

Machine Learning · Computer Science 2018-07-02 Tomaso Poggio , Qianli Liao , Brando Miranda , Andrzej Banburski , Xavier Boix , Jack Hidary

Training a neural network requires navigating a high-dimensional, non-convex loss surface to find parameters that minimize this loss. In many ways, it is surprising that optimizers such as stochastic gradient descent and ADAM can reliably…

Machine Learning · Computer Science 2026-02-06 Conor Rowan , Finn Murphy-Blanchard

Deep neural networks are workhorse models in machine learning with multiple layers of non-linear functions composed in series. Their loss function is highly non-convex, yet empirically even gradient descent minimisation is sufficient to…

Disordered Systems and Neural Networks · Physics 2020-03-18 Simon Becker , Yao Zhang , Alpha A. Lee

Recent work has noted that all bad local minima can be removed from neural network loss landscapes, by adding a single unit with a particular parameterization. We show that the core technique from these papers can be used to remove all bad…

Machine Learning · Statistics 2019-01-15 Jascha Sohl-Dickstein , Kenji Kawaguchi

We analyze recurrent neural networks with diagonal hidden-to-hidden weight matrices, trained with gradient descent in the supervised learning setting, and prove that gradient descent can achieve optimality \emph{without} massive…

Machine Learning · Computer Science 2024-10-11 Semih Cayci , Atilla Eryilmaz

Finding parameters in a deep neural network (NN) that fit training data is a nonconvex optimization problem, but a basic first-order optimization method (gradient descent) finds a global optimizer with perfect fit (zero-loss) in many…

Machine Learning · Computer Science 2025-03-07 Zhiyan Ding , Shi Chen , Qin Li , Stephen Wright

We consider deep linear networks with arbitrary convex differentiable loss. We provide a short and elementary proof of the fact that all local minima are global minima if the hidden layers are either 1) at least as wide as the input layer,…

Machine Learning · Computer Science 2018-07-25 Thomas Laurent , James von Brecht

Solving inverse problems in physics is central to understanding complex systems and advancing technologies in various fields. Iterative optimization algorithms, commonly used to solve these problems, often encounter local minima, chaos, or…

Machine Learning · Computer Science 2025-01-29 Girnar Goyal , Philipp Holl , Sweta Agrawal , Nils Thuerey

Despite their practical success, a theoretical understanding of the loss landscape of neural networks has proven challenging due to the high-dimensional, non-convex, and highly nonlinear structure of such models. In this paper, we…

Machine Learning · Computer Science 2020-07-21 Abbas Kazemipour , Brett W. Larsen , Shaul Druckmann

Neural network training relies on our ability to find "good" minimizers of highly non-convex loss functions. It is well-known that certain network architecture designs (e.g., skip connections) produce loss functions that train easier, and…

Machine Learning · Computer Science 2018-11-08 Hao Li , Zheng Xu , Gavin Taylor , Christoph Studer , Tom Goldstein

Previous theoretical work on deep learning and neural network optimization tend to focus on avoiding saddle points and local minima. However, the practical observation is that, at least in the case of the most successful Deep Convolutional…

Machine Learning · Computer Science 2017-06-23 Qianli Liao , Tomaso Poggio

When training deep neural networks for classification tasks, an intriguing empirical phenomenon has been widely observed in the last-layer classifiers and features, where (i) the class means and the last-layer classifiers all collapse to…

Machine Learning · Computer Science 2022-03-15 Jinxin Zhou , Xiao Li , Tianyu Ding , Chong You , Qing Qu , Zhihui Zhu

We explore some mathematical features of the loss landscape of overparameterized neural networks. A priori one might imagine that the loss function looks like a typical function from $\mathbb{R}^n$ to $\mathbb{R}$ - in particular,…

Machine Learning · Computer Science 2018-04-27 Y Cooper

Under mild assumptions, we investigate the geometry of the loss landscape for two-layer neural networks in the vicinity of global minima. Utilizing novel techniques, we demonstrate: (i) how global minima with zero generalization error…

Machine Learning · Computer Science 2025-04-11 Leyang Zhang , Yaoyu Zhang , Tao Luo

The optimization of neural networks under weight decay remains poorly understood from a theoretical standpoint. While weight decay is standard practice in modern training procedures, most theoretical analyses focus on unregularized…

Machine Learning · Computer Science 2025-05-29 Etienne Boursier , Matthew Bowditch , Matthias Englert , Ranko Lazic

The optimization of multilayer neural networks typically leads to a solution with zero training error, yet the landscape can exhibit spurious local minima and the minima can be disconnected. In this paper, we shed light on this phenomenon:…

Machine Learning · Computer Science 2020-07-24 Alexander Shevchenko , Marco Mondelli

We study the convergence properties of gradient descent for training deep linear neural networks, i.e., deep matrix factorizations, by extending a previous analysis for the related gradient flow. We show that under suitable conditions on…

Machine Learning · Computer Science 2021-11-25 Gabin Maxime Nguegnang , Holger Rauhut , Ulrich Terstiege

By using the viewpoint of modern computational algebraic geometry, we explore properties of the optimization landscapes of the deep linear neural network models. After clarifying on the various definitions of "flat" minima, we show that the…

Machine Learning · Statistics 2018-10-19 Dhagash Mehta , Tianran Chen , Tingting Tang , Jonathan D. Hauenstein

We analyze the optimization landscapes of deep learning with wide networks. We highlight the importance of constraints for such networks and show that constraint -- as well as unconstraint -- empirical-risk minimization over such networks…

Machine Learning · Computer Science 2021-01-14 Johannes Lederer

We systematize the approach to the investigation of deep neural network landscapes by basing it on the geometry of the space of implemented functions rather than the space of parameters. Grouping classifiers into equivalence classes, we…