paper-with-me

홈 › Papers

Beyond Signal Propagation: Is Feature Diversity Necessary in Deep Neural Network Initialization?

2020-07-02 · ICML 2020 1 · Yaniv Blumenfeld, Dar Gilboa, Daniel Soudry

Deep neural networks are typically initialized with random weights, with variances chosen to facilitate signal propagation and stable gradients. It is also believed that diversity of features is an important property of these initializations. We construct a deep convolutional network with identical features by initializing almost all the weights to $0$. The architecture also enables perfect signal propagation and stable gradients, and achieves high accuracy on standard benchmarks. This indicates that random, diverse initializations are \textit{not} necessary for training neural networks. An essential element in training this network is a mechanism of symmetry breaking; we study this phenomenon and find that standard GPU operations, which are non-deterministic, can serve as a sufficient source of symmetry breaking to enable training.

📄 PDF Abstract BibTeX arXiv:2007.01038

Code (1)

yanivbl6/BeyondSigProp 공식 구현 pytorch

Tasks

DiversityGPU

Similar Papers 제목 키워드 기반

Signal Propagation: A Framework for Learning and Inference In a Forward Pass

2022-04-04 · Adam Kohan, Edward A. Rietman, Hava T. Siegelmann

We propose a new learning framework, signal propagation (sigprop), for propagating a learning signal and updating neural network parameters via a forward pass, as an alternative to backpropagation. In sigprop, there is o…

Does Graph Compression Preserve Signal Propagation?

2026-07-25 · Kawshik Banerjee, Khaled Mohammed Saifuddin arxiv

Graph compression reduces the computational cost of graph learning, but its effect on signal propagation remains largely underexplored. Existing work evaluates compression through downstream task performance or structura…

Graph Learning

Biologically-plausible backpropagation through arbitrary timespans via local neuromodulators

2022-06-02 · Yuhan Helena Liu, Stephen Smith, Stefan Mihalas, Eric Shea-Brown 외

The spectacular successes of recurrent neural network models where key parameters are adjusted via backpropagation-based gradient descent have inspired much thought as to how biological neuronal networks might solve the …

Effective Resistance Rewiring: A Simple Topological Correction for Over-Squashing

2026-03-12 · Bertran Miquel-Oliver, Manel Gil-Sorribes, Victor Guallar, Alexis Molina arxiv

Graph Neural Networks struggle to capture long-range dependencies due to over-squashing, where information from exponentially growing neighborhoods must pass through a small number of structural bottlenecks. While recent…

Beyond Oversquashing: Understanding Signal Propagation in GNNs Via Observables

2026-05-13 · Eden Nagar, Ya-Wei Eileen Lin, Ron Levie arxiv

Graph Neural Networks (GNNs) perform computations on graphs by routing the signal between graph regions using a graph shift operator or a message passing scheme. Often, the propagation of the signal leads to a loss of in…