English
Related papers

Related papers: Do Deep Networks Forget Initialization? A Forgetti…

200 papers

In this work, we investigate the use of sparsity-inducing regularizers during training of Convolution Neural Networks (CNNs). These regularizers encourage that fewer connections in the convolution and fully connected layers take non-zero…

Computer Vision and Pattern Recognition · Computer Science 2014-12-04 Maxwell D. Collins , Pushmeet Kohli

Adam is the de facto optimization algorithm for several deep learning applications, but an understanding of its implicit bias and how it differs from other algorithms, particularly standard first-order methods such as (stochastic) gradient…

Machine Learning · Computer Science 2025-10-27 Bhavya Vasudeva , Jung Whan Lee , Vatsal Sharan , Mahdi Soltanolkotabi

Models for learning probability distributions such as generative models and density estimators behave quite differently from models for learning functions. One example is found in the memorization phenomenon, namely the ultimate convergence…

Machine Learning · Statistics 2021-03-03 Hongkang Yang , Weinan E

Overparameterised deep neural networks (DNNs) are highly expressive and so can, in principle, generate almost any function that fits a training dataset with zero error. The vast majority of these functions will perform poorly on unseen…

Machine Learning · Computer Science 2021-04-13 Chris Mingard , Guillermo Valle-Pérez , Joar Skalse , Ard A. Louis

Training neural networks with first order optimisation methods is at the core of the empirical success of deep learning. The scale of initialisation is a crucial factor, as small initialisations are generally associated to a feature…

Machine Learning · Computer Science 2025-09-16 Etienne Boursier , Nicolas Flammarion

Regardless of the marked differences between biological and artificial neural systems, one fundamental similarity is that they are essentially dynamical systems that can learn to imitate other dynamical systems, without knowing their…

Neurons and Cognition · Quantitative Biology 2019-11-06 Zhixin Lu , Danielle S. Bassett

We present a non-asymptotic theory of generalization in deep learning where the empirical neural tangent kernel partitions the output space. In directions corresponding to signal, error dissipates rapidly; in the vast orthogonal dimensions…

Machine Learning · Computer Science 2026-05-05 Elon Litman , Gabe Guo

Diffusion probabilistic models have become a cornerstone of modern generative AI, yet the mechanisms underlying their generalization remain poorly understood. In fact, if these models were perfectly minimizing their training loss, they…

Machine Learning · Computer Science 2025-09-03 Alessandro Favero , Antonio Sclocchi , Matthieu Wyart

Modern deep networks have millions to billions of parameters, which leads to high memory and energy requirements during training as well as during inference on resource-constrained edge devices. Consequently, pruning techniques have been…

Machine Learning · Computer Science 2020-03-06 Sourjya Roy , Priyadarshini Panda , Gopalakrishnan Srinivasan , Anand Raghunathan

Standard practice in training neural networks involves initializing the weights in an independent fashion. The results of recent work suggest that feature "diversity" at initialization plays an important role in training the network.…

Machine Learning · Computer Science 2020-07-06 Yaniv Blumenfeld , Dar Gilboa , Daniel Soudry

Training vision transformer networks on small datasets poses challenges. In contrast, convolutional neural networks (CNNs) can achieve state-of-the-art performance by leveraging their architectural inductive bias. In this paper, we…

Computer Vision and Pattern Recognition · Computer Science 2024-01-24 Jianqiao Zheng , Xueqian Li , Simon Lucey

Current deep learning architectures suffer from catastrophic forgetting, a failure to retain knowledge of previously learned classes when incrementally trained on new classes. The fundamental roadblock faced by deep learning methods is that…

Machine Learning · Computer Science 2020-12-01 Ziyang Wu , Christina Baek , Chong You , Yi Ma

The infrequent occurrence of overfit in deep neural networks is perplexing. On the one hand, theory predicts that as models get larger they should eventually become too specialized for a specific training set, with ensuing decrease in…

Machine Learning · Computer Science 2023-12-29 Uri Stern , Daphna Weinshall

State-of-the-art deep neural networks are still struggling to address the catastrophic forgetting problem in continual learning. In this paper, we propose one simple paradigm (named as S-Prompting) and two concrete approaches to highly…

Computer Vision and Pattern Recognition · Computer Science 2023-03-21 Yabin Wang , Zhiwu Huang , Xiaopeng Hong

We demonstrate that in residual neural networks (ResNets) dynamical isometry is achievable irrespectively of the activation function used. We do that by deriving, with the help of Free Probability and Random Matrix Theories, a universal…

Machine Learning · Statistics 2019-03-05 Wojciech Tarnowski , Piotr Warchoł , Stanisław Jastrzębski , Jacek Tabor , Maciej A. Nowak

We show that the influence of a subset of the training samples can be removed -- or "forgotten" -- from the weights of a network trained on large-scale image classification tasks, and we provide strong computable bounds on the amount of…

Machine Learning · Computer Science 2021-06-22 Aditya Golatkar , Alessandro Achille , Avinash Ravichandran , Marzia Polito , Stefano Soatto

Sparsity in the structure of Neural Networks can lead to less energy consumption, less memory usage, faster computation times on convenient hardware, and automated machine learning. If sparsity gives rise to certain kinds of structure, it…

Machine Learning · Computer Science 2021-07-28 Julian Stier , Harshil Darji , Michael Granitzer

We conjecture that the inherent difference in generalisation between adaptive and non-adaptive gradient methods in deep learning stems from the increased estimation noise in the flattest directions of the true loss surface. We demonstrate…

Machine Learning · Statistics 2022-03-17 Diego Granziol , Nicholas Baskerville

In many real-world deployments of machine learning systems, data arrive piecemeal. These learning scenarios may be passive, where data arrive incrementally due to structural properties of the problem (e.g., daily financial data) or active,…

Machine Learning · Computer Science 2021-01-01 Jordan T. Ash , Ryan P. Adams

Biological neural networks are capable of recruiting different sets of neurons to encode different memories. However, when training artificial neural networks on a set of tasks, typically, no mechanism is employed for selectively producing…

Machine Learning · Computer Science 2023-05-17 Matthew J. Tilley , Michelle Miller , David J. Freedman