Zobrazeno 1 - 5
of 5
pro vyhledávání: '"Negiar, Geoffrey"'
Autor:
Olivares, Kin G., Négiar, Geoffrey, Ma, Ruijun, Meetei, O. Nangba, Cao, Mengfei, Mahoney, Michael W.
Obtaining accurate probabilistic forecasts is an important operational challenge in many applications, like energy management, climate forecast, supply chain planning, and resource allocation. In many of these applications, there is a natural hierarc
Externí odkaz:
http://arxiv.org/abs/2307.09797
We introduce a practical method to enforce partial differential equation (PDE) constraints for functions defined by neural networks (NNs), with a high degree of accuracy and up to a desired tolerance. We develop a differentiable PDE-constrained layer
Externí odkaz:
http://arxiv.org/abs/2207.08675
Autor:
Négiar, Geoffrey, Dresdner, Gideon, Tsai, Alicia, Ghaoui, Laurent El, Locatello, Francesco, Freund, Robert M., Pedregosa, Fabian
We propose a novel Stochastic Frank-Wolfe (a.k.a. conditional gradient) algorithm for constrained smooth finite-sum minimization with a generalized linear prediction/structure. This class of problems includes empirical risk minimization with sparse,
Externí odkaz:
http://arxiv.org/abs/2002.11860
Publikováno v:
Proceedings of the 23rd International Conference on Artificial Intelligence and Statistics (AISTATS) 2020
Structured constraints in Machine Learning have recently brought the Frank-Wolfe (FW) family of algorithms back in the spotlight. While the classical FW algorithm has poor local convergence properties, the Away-steps and Pairwise FW variants have eme
Externí odkaz:
http://arxiv.org/abs/1806.05123
We describe a novel family of models of multi- layer feedforward neural networks in which the activation functions are encoded via penalties in the training problem. Our approach is based on representing a non-decreasing activation function as the ar
Externí odkaz:
http://arxiv.org/abs/1805.01532