Zobrazeno 1 - 10
of 81
pro vyhledávání: '"Roosta, Fred"'
Autor:
Smee, Oscar, Roosta, Fred
We consider solving large scale nonconvex optimisation problems with nonnegativity constraints. Such problems arise frequently in machine learning, such as nonnegative least-squares, nonnegative matrix factorisation, as well as problems with sparsity
Externí odkaz:
http://arxiv.org/abs/2405.12401
In this paper, we dive into the reliability concerns of Integrated Gradients (IG), a prevalent feature attribution method for black-box deep learning models. We particularly address two predominant challenges associated with IG: the generation of noi
Externí odkaz:
http://arxiv.org/abs/2405.09800
In the realm of medical imaging, inverse problems aim to infer high-quality images from incomplete, noisy measurements, with the objective of minimizing expenses and risks to patients in clinical settings. The Diffusion Models have recently emerged a
Externí odkaz:
http://arxiv.org/abs/2404.03706
Since the development of the conjugate gradient (CG) method in 1952 by Hestenes and Stiefel, CG, has become an indispensable tool in computational mathematics for solving positive definite linear systems. On the other hand, the conjugate residual (CR
Externí odkaz:
http://arxiv.org/abs/2401.11714
We develop a new efficient sequential approximate leverage score algorithm, SALSA, using methods from randomized numerical linear algebra (RandNLA) for large matrices. We demonstrate that, with high probability, the accuracy of SALSA's approximations
Externí odkaz:
http://arxiv.org/abs/2401.00122
The analysis of gradient descent-type methods typically relies on the Lipschitz continuity of the objective gradient. This generally requires an expensive hyperparameter tuning process to appropriately calibrate a stepsize for a given problem. In thi
Externí odkaz:
http://arxiv.org/abs/2311.08615
Deep learning is renowned for its theory-practice gap, whereby principled theory typically fails to provide much beneficial guidance for implementation in practice. This has been highlighted recently by the benign overfitting phenomenon: when neural
Externí odkaz:
http://arxiv.org/abs/2311.07013
The celebrated minimum residual method (MINRES), proposed in the seminal paper of Paige and Saunders, has seen great success and wide-spread use in solving linear least-squared problems involving Hermitian matrices, with further extensions to complex
Externí odkaz:
http://arxiv.org/abs/2309.17096
Autor:
Lim, Alexander, Roosta, Fred
We consider extensions of the Newton-MR algorithm for nonconvex optimization to the settings where Hessian information is approximated. Under additive noise model on the Hessian matrix, we investigate the iteration and operation complexities of these
Externí odkaz:
http://arxiv.org/abs/2308.09912
The problem of model selection is considered for the setting of interpolating estimators, where the number of model parameters exceeds the size of the dataset. Classical information criteria typically consider the large-data limit, penalizing model s
Externí odkaz:
http://arxiv.org/abs/2307.07785