linear networks - Robuta Search

https://openreview.net/forum?id=F738WY1Xm4&referrer=%5Bthe%20profile%20of%20Pierre%20Marion%5D(%2Fprofile%3Fid%3D~Pierre_Marion1) Deep linear networks for regression are implicitly regularized towards flat minima | OpenReview The largest eigenvalue of the Hessian, or sharpness, of neural networks is a key quantity to understand their optimization dynamics. In this paper, we study... linear networks https://openreview.net/forum?id=jjpsFetXJp Neural Collapse in Deep Linear Networks: From Balanced to Imbalanced Data | OpenReview Modern deep neural networks have achieved impressive performance on tasks from image classification to natural language processing. Surprisingly, these complex... in deep linear networks https://openreview.net/forum?id=TSaieShX3j SGD vs GD: Rank Deficiency in Linear Networks | OpenReview In this article, we study the behaviour of continuous-time gradient methods on a two-layer linear network with square loss. A dichotomy between SGD and GD is... rank deficiency linear networks sgd vs openreview https://openreview.net/forum?id=O6znYvxC1U Bayesian Treatment of the Spectrum of the Empirical Kernel in (Sub)Linear-Width Neural Networks |... We study Bayesian neural networks (BNNs) in the theoretical limits of infinitely increasing number of training examples, network width and input space... https://www.thewrap.com/amc-networks-earnings-q3-2025/ AMC Networks US Ad Revenue Drops 17% on Linear TV Decline Nov 7, 2025 - Affiliate revenues also tumbled 13%, but streaming was a bright spot as subscribers grew 2% to 10.4 million amc networks ad revenue linear tv us https://deepai.org/publication/geometry-of-linear-convolutional-networks Geometry of Linear Convolutional Networks | DeepAI Aug 3, 2021 - 08/03/21 - We study the family of functions that are represented by a linear convolutional neural network (LCN). These functions form a semi-... geometry linear networks deepai https://www.mediapost.com/publications/article/407867/legacy-headache-big-tech-not-buying-linear-tv-net.html Legacy Headache: Big Tech Not Buying Linear TV Networks Anytime Soon 08/04/2025 Legacy Headache: Big Tech Not Buying Linear TV Networks Anytime Soon - 08/04/2025 https://arxiv.org/abs/1501.01376v1 [1501.01376v1] Secure Transmission in Wireless Sensor Networks Data Using Linear Kolmogorov... Abstract page for arXiv paper 1501.01376v1: Secure Transmission in Wireless Sensor Networks Data Using Linear Kolmogorov Watermarking Technique wireless sensor networks secure transmission https://arxiv.org/abs/2505.03577 [2505.03577] Information-theoretic reduction of deep neural networks to linear models in the... Abstract page for arXiv paper 2505.03577: Information-theoretic reduction of deep neural networks to linear models in the overparametrized proportional regime https://openreview.net/forum?id=B1x6BTEKwr Piecewise linear activations substantially shape the loss surfaces of neural networks | OpenReview This paper presents how the loss surfaces of nonlinear neural networks are substantially shaped by the nonlinearities in activations. piecewise linear the loss https://deepai.org/publication/computing-in-anonymous-dynamic-networks-is-linear Computing in Anonymous Dynamic Networks Is Linear | DeepAI Apr 5, 2022 - 04/05/22 - We give the first linear-time counting algorithm for processes in anonymous 1-interval-connected dynamic networks with a leader. A... dynamic networks computing anonymous linear deepai https://openreview.net/forum?id=DALK4KJTjX What Makes Freezing Layers in Deep Neural Networks Effective? A Linear Separability Perspective |... Freezing layers in deep neural networks has been shown to enhance generalization and accelerate training, yet the underlying mechanisms remain unclear. This... deep neural networks https://openreview.net/forum?id=SysEexbRb Critical Points of Linear Neural Networks: Analytical Forms and Landscape Properties | OpenReview We provide necessary and sufficient analytical forms for the critical points of the square loss functions for various neural networks, and exploit the... critical points neural networks https://arxiv.org/abs/2211.08516 [2211.08516] Phenotype Search Trajectory Networks for Linear Genetic Programming Abstract page for arXiv paper 2211.08516: Phenotype Search Trajectory Networks for Linear Genetic Programming 2211 phenotype search trajectory networks https://jmlr.org/papers/v23/21-0368.html Approximation and Optimization Theory for Linear Continuous-Time Recurrent Neural Networks optimization theory continuous time approximation https://openreview.net/forum?id=pcpjtYNJCH Initialization-Dependent Sample Complexity of Linear Predictors and Neural Networks | OpenReview We provide several new results on the sample complexity of vector-valued linear predictors (parameterized by a matrix), and more generally neural networks.... sample complexity neural networks initialization dependent https://openreview.net/forum?id=Su-G6VpE1W5 Compositional Multi-Object Reinforcement Learning with Linear Relation Networks | OpenReview Although reinforcement learning has seen remarkable progress over the last years, solving robust dexterous object-manipulation tasks in multi-object settings... reinforcement learning linear relation compositional multi object https://deepai.org/publication/inverted-residuals-and-linear-bottlenecks-mobile-networks-forclassification-detection-and-segmentation Inverted Residuals and Linear Bottlenecks: Mobile Networks forClassification, Detection and... Jan 13, 2018 - 01/13/18 - In this paper we describe a new mobile architecture, MobileNetV2, that improves the state of the art performance of mobile models ... mobile networks inverted residuals linear bottlenecks https://openreview.net/forum?id=NQ8YNyX2uy The Computational Complexity of Counting Linear Regions in ReLU Neural Networks | OpenReview An established measure of the expressive power of a given ReLU neural network is the number of linear regions into which it partitions the input space. There... computational complexity https://openreview.net/forum?id=Yg7tt1hWiF A Lower Bound for the Number of Linear Regions of Ternary ReLU Regression Neural Networks |... With the advancement of deep learning, reducing computational complexity and memory consumption has become a critical challenge, and ternary neural networks... https://openreview.net/forum?id=uAyElhYKxg&referrer=%5Bthe%20profile%20of%20Mathieu%20Even%5D(%2Fprofile%3Fid%3D~Mathieu_Even1) (S)GD over Diagonal Linear Networks: Implicit bias, Large Stepsizes and Edge of Stability |... In this paper, we investigate the impact of stochasticity and large stepsizes on the implicit regularisation of gradient descent (GD) and stochastic gradient... https://openreview.net/forum?id=B1J_rgWRW Understanding Deep Neural Networks with Rectified Linear Units | OpenReview This paper 1) characterizes functions representable by ReLU DNNs, 2) formally studies the benefit of depth in such architectures, 3) gives an algorithm to... deep neural networks linear units understanding rectified openreview https://deepai.org/publication/reverse-engineering-recurrent-neural-networks-with-jacobian-switching-linear-dynamical-systems Reverse engineering recurrent neural networks with Jacobian switching linear dynamical systems |... Nov 1, 2021 - 11/01/21 - Recurrent neural networks (RNNs) are powerful models for processing time-series data, but it remains challenging to understand how... recurrent neural networks reverse engineering https://openreview.net/forum?id=S9kFcPHqHP Critical Points and Convergence Analysis of Generative Deep Linear Networks Trained with... We consider a deep matrix factorization model of covariance matrices trained with the Bures-Wasserstein distance. While recent works have made advances in the... critical points https://www.sintef.no/en/publications/publication/2196398/ Hierarchical Analytical Marching of Piecewise Linear Neural Networks - SINTEF piecewise linear neural networks hierarchical analytical marching https://arxiv.org/abs/1801.04381v1 [1801.04381v1] Inverted Residuals and Linear Bottlenecks: Mobile Networks forClassification,... Abstract page for arXiv paper 1801.04381v1: Inverted Residuals and Linear Bottlenecks: Mobile Networks forClassification, Detection and Segmentation mobile networks 1801 inverted residuals linear https://deepai.org/publication/locally-refined-quad-meshing-for-linear-elasticity-problems-based-on-convolutional-neural-networks Locally refined quad meshing for linear elasticity problems based on convolutional neural networks... Mar 15, 2022 - 03/15/22 - In this paper we propose a method to generate suitably refined finite element meshes using neural networks. As a model problem we ... https://openreview.net/forum?id=D2Oaj7v9YJ LinSATNet: The Positive Linear Satisfiability Neural Networks | OpenReview Encoding constraints into neural networks is attractive. This paper studies how to introduce the popular positive linear satisfiability to neural networks. We... the positive neural networks linear satisfiability openreview https://deepai.org/publication/attributions-beyond-neural-networks-the-linear-program-case Attributions Beyond Neural Networks: The Linear Program Case | DeepAI Jun 14, 2022 - 06/14/22 - Linear Programs (LPs) have been one of the building blocks in machine learning and have championed recent strides in differentiabl... neural networks linear program attributions beyond case https://openreview.net/forum?id=gpqBGyKeKH Spectral Evolution and Invariance in Linear-width Neural Networks | OpenReview We investigate the spectral properties of linear-width feed-forward neural networks, where the sample size is asymptotically proportional to network width.... neural networks spectral evolution invariance linear https://openreview.net/forum?id=_wzZwKpTDF_9C Exact solutions to the nonlinear dynamics of learning in deep linear neural networks | OpenReview Despite the widespread practical success of deep learning methods, our theoretical understanding of the dynamics of learning in deep neural networks remains... https://deepai.org/publication/pure-and-spurious-critical-points-a-geometric-study-of-linear-networks Pure and Spurious Critical Points: a Geometric Study of Linear Networks | DeepAI Oct 3, 2019 - 10/03/19 - The critical locus of the loss function of a neural network is determined by the geometry of the functional space and by the param... critical points