paper-with-me

Papers

Tractable structured natural gradient descent using local parameterizations

2021-02-15 · Wu Lin, Frank Nielsen, Mohammad Emtiyaz Khan, Mark Schmidt

Natural-gradient descent (NGD) on structured parameter spaces (e.g., low-rank covariances) is computationally challenging due to difficult Fisher-matrix computations. We address this issue by using \emph{local-parameter coordinates} to obtain a flexible and efficient NGD method that works well for a wide-variety of structured parameterizations. We show four applications where our method (1) generalizes the exponential natural evolutionary strategy, (2) recovers existing Newton-like algorithms, (3) yields new structured second-order algorithms via matrix groups, and (4) gives new algorithms to learn covariances of Gaussian and Wishart-based distributions. We show results on a range of problems from deep learning, variational inference, and evolution strategies. Our work opens a new direction for scalable structured geometric methods.

📄 PDF Abstract BibTeX arXiv:2102.07405

Code (0)

등록된 구현이 없습니다.

Tasks

Variational Inference

Similar Papers 제목 키워드 기반

Bethe Projections for Non-Local Inference

2015-03-04 · Luke Vilnis, David Belanger, Daniel Sheldon, Andrew McCallum

Many inference problems in structured prediction are naturally solved by augmenting a tractable dependency structure with complex, non-local auxiliary objectives. This includes the mean field family of variational infere…

Handwriting RecognitionStructured PredictionVariational Inference

Structured second-order methods via natural gradient descent

2021-07-22 · Wu Lin, Frank Nielsen, Mohammad Emtiyaz Khan, Mark Schmidt

In this paper, we propose new structured second-order methods and structured adaptive-gradient methods obtained by performing natural-gradient descent on structured parameter spaces. Natural-gradient descent is an attrac…

Second-order methods

On Tractable $Φ$-Equilibria in Non-Concave Games

2024-03-13 · Yang Cai, Constantinos Daskalakis, Haipeng Luo, Chen-Yu Wei 외

While Online Gradient Descent and other no-regret learning procedures are known to efficiently converge to a coarse correlated equilibrium in games where each agent's utility is concave in their own strategy, this is not…

A Kronecker-factored approximate Fisher matrix for convolution layers

2016-02-03 · Roger Grosse, James Martens

Second-order optimization methods such as natural gradient descent have the potential to speed up training of neural networks by correcting for the curvature of the loss function. Unfortunately, the exact natural gradien…

Stochastic Optimization

Reconstructing Deep Neural Networks: Unleashing the Optimization Potential of Natural Gradient Descent

2024-12-10 · Weihua Liu, Said Boumaraf, Jianwu Li, Chaochao Lin 외

Natural gradient descent (NGD) is a powerful optimization technique for machine learning, but the computational complexity of the inverse Fisher information matrix limits its application in training deep neural networks.…