paper-with-me

Papers

Implicit Regularization via Neural Feature Alignment

2020-08-03 · NeurIPS Workshop DL-IG 2020 12 · Aristide Baratin, Thomas George, César Laurent, R. Devon Hjelm, Guillaume Lajoie, Pascal Vincent, Simon Lacoste-Julien

We approach the problem of implicit regularization in deep learning from a geometrical viewpoint. We highlight a regularization effect induced by a dynamical alignment of the neural tangent features introduced by Jacot et al, along a small number of task-relevant directions. This can be interpreted as a combined mechanism of feature selection and compression. By extrapolating a new analysis of Rademacher complexity bounds for linear models, we motivate and study a heuristic complexity measure that captures this phenomenon, in terms of sequences of tangent kernel classes along optimization paths.

📄 PDF Abstract BibTeX arXiv:2008.00938

Code (1)

tfjgeorge/ntk_alignment 공식 구현 pytorch

Tasks

feature selectionModel Compression

Methods 이 논문이 사용한 방법론

Feature Selection Feature selection, also known as variable selection, attribute selection or variable subset selection, is the process of selecting a subset of relevant features (variables,…

Similar Papers 제목 키워드 기반

Implicit Gradient Alignment in Distributed and Federated Learning

2021-06-25 · Yatin Dandi, Luis Barba, Martin Jaggi

A major obstacle to achieving global convergence in distributed and federated learning is the misalignment of gradients across clients, or mini-batches due to heterogeneity and stochasticity of the distributed data. In t…

Federated Learning

Convergence Analysis and Implicit Regularization of Feedback Alignment for Deep Linear Networks

2021-10-20 · Manuela Girotti, Ioannis Mitliagkas, Gauthier Gidel

We theoretically analyze the Feedback Alignment (FA) algorithm, an efficient alternative to backpropagation for training neural networks. We provide convergence guarantees with rates for deep linear networks for both con…

Incremental Learning

Implicit Regularization of Gradient Flow on One-Layer Softmax Attention

2024-03-13 · Heejune Sheen, Siyu Chen, Tianhao Wang, Harrison H. Zhou

We study gradient flow on the exponential loss for a classification problem with a one-layer softmax attention model, where the key and query weight matrices are trained separately. Under a separability assumption on the…

SpaceJAM: a Lightweight and Regularization-free Method for Fast Joint Alignment of Images

2024-07-16 · Nir Barel, Ron Shapira Weber, Nir Mualem, Shahaf E. Finder 외

The unsupervised task of Joint Alignment (JA) of images is beset by challenges such as high complexity, geometric distortions, and convergence to poor local or even global optima. Although Vision Transformers (ViT) have …

16k

Implicit Regularization in Feedback Alignment Learning Mechanisms for Neural Networks

2023-06-02 · Zachary Robertson, Oluwasanmi Koyejo

Feedback Alignment (FA) methods are biologically inspired local learning rules for training neural networks with reduced communication between layers. While FA has potential applications in distributed and privacy-aware …

Multi-class Classification