English
Related papers

Related papers: Revisiting Initialization of Neural Networks

200 papers

In gradient descent dynamics of neural networks, the top eigenvalue of the loss Hessian (sharpness) displays a variety of robust phenomena throughout training. This includes early time regimes where the sharpness may decrease during early…

Machine Learning · Computer Science 2025-02-17 Dayal Singh Kalra , Tianyu He , Maissam Barkeshli

Echo State Networks (ESNs) are a special type of recurrent neural networks (RNNs), in which the input and recurrent connections are traditionally generated randomly, and only the output weights are trained. Despite the recent success of…

Machine Learning · Computer Science 2022-05-11 Peter Steiner , Azarakhsh Jalalvand , Peter Birkholz

We propose a simultaneous learning and pruning algorithm capable of identifying and eliminating irrelevant structures in a neural network during the early stages of training. Thus, the computational cost of subsequent training iterations,…

Machine Learning · Computer Science 2023-01-16 Valentin Frank Ingmar Guenter , Athanasios Sideris

Regularization is typically understood as improving generalization by altering the landscape of local extrema to which the model eventually converges. Deep neural networks (DNNs), however, challenge this view: We show that removing…

Machine Learning · Computer Science 2019-06-03 Aditya Golatkar , Alessandro Achille , Stefano Soatto

Deep neural networks achieve state-of-the-art performance for a range of classification and inference tasks. However, the use of stochastic gradient descent combined with the nonconvexity of the underlying optimization problems renders…

Machine Learning · Computer Science 2020-01-29 Ramina Ghods , Andrew S. Lan , Tom Goldstein , Christoph Studer

Overparameterized Neural Networks (NN) display state-of-the-art performance. However, there is a growing need for smaller, energy-efficient, neural networks tobe able to use machine learning applications on devices with limited…

Machine Learning · Statistics 2021-05-21 Soufiane Hayou , Jean-Francois Ton , Arnaud Doucet , Yee Whye Teh

Recent work has explored the possibility of pruning neural networks at initialization. We assess proposals for doing so: SNIP (Lee et al., 2019), GraSP (Wang et al., 2020), SynFlow (Tanaka et al., 2020), and magnitude pruning. Although…

Machine Learning · Computer Science 2021-03-23 Jonathan Frankle , Gintare Karolina Dziugaite , Daniel M. Roy , Michael Carbin

Neural networks have achieved remarkable success in many cognitive tasks. However, when they are trained sequentially on multiple tasks without access to old data, their performance on early tasks tend to drop significantly. This problem is…

Machine Learning · Computer Science 2021-02-10 Dong Yin , Mehrdad Farajtabar , Ang Li , Nir Levine , Alex Mott

Supervised training of deep neural nets typically relies on minimizing cross-entropy. However, in many domains, we are interested in performing well on metrics specific to the application. In this paper we propose a direct loss minimization…

Machine Learning · Computer Science 2016-06-03 Yang Song , Alexander G. Schwing , Richard S. Zemel , Raquel Urtasun

Humans learn continually throughout their lifespan by accumulating diverse knowledge and fine-tuning it for future tasks. When presented with a similar goal, neural networks suffer from catastrophic forgetting if data distributions across…

Machine Learning · Computer Science 2022-09-19 Dupati Srikar Chandra , Sakshi Varshney , P. K. Srijith , Sunil Gupta

This paper proposes a sensitivity analysis framework based on set valued mapping for deep neural networks (DNN) to understand and compute how the solutions (model weights) of DNN respond to perturbations in the training data. As a DNN may…

Machine Learning · Computer Science 2024-12-17 Xin Wang , Feilong Wang , Xuegang Ban

Stochastic gradient descent (SGD) has been the dominant optimization method for training deep neural networks due to its many desirable properties. One of the more remarkable and least understood quality of SGD is that it generalizes…

Machine Learning · Computer Science 2020-07-03 Erhan Bilal

Optimizing deep neural networks (DNNs) often suffers from the ill-conditioned problem. We observe that the scaling-based weight space symmetry property in rectified nonlinear network will cause this negative effect. Therefore, we propose to…

Machine Learning · Computer Science 2017-10-09 Lei Huang , Xianglong Liu , Bo Lang , Bo Li

Threshold activation functions are highly preferable in neural networks due to their efficiency in hardware implementations. Moreover, their mode of operation is more interpretable and resembles that of biological neurons. However,…

Machine Learning · Computer Science 2023-03-07 Tolga Ergen , Halil Ibrahim Gulluk , Jonathan Lacotte , Mert Pilanci

Enforcing orthonormal or isometric property for the weight matrices has been shown to enhance the training of deep neural networks by mitigating gradient exploding/vanishing and increasing the robustness of the learned networks. However,…

Machine Learning · Computer Science 2024-03-01 Zhen Qin , Xuwei Tan , Zhihui Zhu

Even though dense networks have lost importance today, they are still used as final logic elements. It could be shown that these dense networks can be simplified by the sparse graph interpretation. This in turn shows that the information…

Neural and Evolutionary Computing · Computer Science 2018-09-25 Thomas Pircher , Dominik Haspel , Eberhard Schlücker

Proper initialisation strategy is of primary importance to mitigate gradient explosion or vanishing when training neural networks. Yet, the impact of initialisation parameters still lacks a precise theoretical understanding for several…

Machine Learning · Computer Science 2026-05-12 Andrea Combette , Antoine Venaille , Nelly Pustelnik

Training a neural network (NN) depends on multiple factors, including but not limited to the initial weights. In this paper, we focus on initializing deep NN parameters such that it performs better, comparing to random or zero…

Machine Learning · Computer Science 2020-11-10 Mohamad H. Danesh

To understand the dynamics of optimization in deep neural networks, we develop a tool to study the evolution of the entire Hessian spectrum throughout the optimization process. Using this, we study a number of hypotheses concerning…

Machine Learning · Computer Science 2019-01-30 Behrooz Ghorbani , Shankar Krishnan , Ying Xiao

Recent results suggest that reinitializing a subset of the parameters of a neural network during training can improve generalization, particularly for small training sets. We study the impact of different reinitialization methods in several…

Machine Learning · Computer Science 2021-09-02 Ibrahim Alabdulmohsin , Hartmut Maennel , Daniel Keysers