Global covariance pooling (GCP) aims at exploiting the second-order statistics of the convolutional feature. Its effectiveness has been demonstrated in boosting the classification performance of Convolutional Neural Networks (CNNs). Singular Value Decomposition (SVD) is used in GCP to compute the matrix square root. However, the approximate matrix square root calculated using Newton-Schulz iteration \cite{li2018towards} outperforms the accurate one computed via SVD \cite{li2017second}. We emp...| papers.cool
Eigendecomposition of symmetric matrices is at the heart of many computer vision algorithms. However, the derivatives of the eigenvectors tend to be numerically unstable, whether using the SVD to compute them analytically or using the Power Iteration (PI) method to approximate them. This instability arises in the presence of eigenvalues that are close to each other. This makes integrating eigendecomposition into deep networks difficult and often results in poor convergence, particularly when ...| papers.cool
Eigendecomposition (ED) is widely used in deep networks. However, the backpropagation of its results tends to be numerically unstable, whether using ED directly or approximating it with the Power Iteration method, particularly when dealing with large matrices. While this can be mitigated by partitioning the data in small and arbitrary groups, doing so has no theoretical basis and makes its impossible to exploit the power of ED to the full. In this paper, we introduce a numerically stable and ...| papers.cool
Deep neural network architectures have recently produced excellent results in a variety of areas in artificial intelligence and visual recognition, well surpassing traditional shallow architectures trained using hand-designed features. The power of deep networks stems both from their ability to perform local computations followed by pointwise non-linearities over increasingly larger receptive fields, and from the simplicity and scalability of the gradient-descent training procedure based on b...| openaccess.thecvf.com
在文章《Muon优化器赏析：从向量到矩阵的本质跨越》中，我们介绍了一个名为“Muon”的新优化器，其中一个理解视角是作为谱范数正则下的最速梯度下降，这似乎揭示了矩阵参数的更本质的优化方向。众所周...| kexue.fm
可能很多读者跟笔者一样，对矩阵的低秩近似有种熟悉而又陌生的感觉。熟悉是因为，低秩近似的概念和意义都不难理解，加之目前诸如LoRA等基于低秩近似的微调技术遍地开花，让低秩近似的概念在耳濡目染间就已...| kexue.fm
随着LLM时代的到来，学术界对于优化器的研究热情似乎有所减退。这主要是因为目前主流的AdamW已经能够满足大多数需求，而如果对优化器“大动干戈”，那么需要巨大的验证成本。因此，当前优化器的变化，...| kexue.fm
上一篇文章中我们介绍了“伪逆”，它关系到给定矩阵$\boldsymbol{M}$和$\boldsymbol{A}$（或$\boldsymbol{B}$）时优化目标$\Vert \boldsymb...| kexue.fm
SVD（Singular Value Decomposition，奇异值分解）是常见的矩阵分解算法，相信很多读者都已经对它有所了解，此前我们在《低秩近似之路（二）：SVD》也专门介绍过它。然而，...| kexue.fm
事实上，除了写博客内容，在这几年里，笔者是花了相当一部分时间来做科学空间的“表面功夫”，为此还专门学了一点php、css和js。虽然不敢说精益求精，但总体来说网站的浏览体验应该比前几年要好得多。...| kexue.fm