Results 71 to 80 of about 4,609,632 (195)
Singular Bayesian Neural Networks
Bayesian neural networks promise calibrated uncertainty but require $O(mn)$ parameters for standard mean-field Gaussian posteriors. We argue this cost is often unnecessary, particularly when weight matrices exhibit fast singular value decay. By parameterizing weights as $W = AB^{\top}$ with $A \in \mathbb{R}^{m \times r}$, $B \in \mathbb{R}^{n \times r}
Mame Diarra Toure, David A. Stephens
openaire +3 more sources
Variational Laplace for Bayesian neural networks
We develop variational Laplace for Bayesian neural networks (BNNs) which exploits a local approximation of the curvature of the likelihood to estimate the ELBO without the need for stochastic sampling of the neural-network weights. The Variational Laplace objective is simple to evaluate, as it is (in essence) the log-likelihood, plus weight-decay, plus
Unlu, Ali, Aitchison, Laurence
openaire +4 more sources
Probabilistic classification of quality of service in wireless computer networks
There is an increasing reliance on wireless computer networks for communicating various types of time sensitive applications such as voice over internet protocol (VoIP). Quality of service (QoS) can play an important role in wireless computer networks as
Abdussalam Salama, Reza Saatchi
doaj +1 more source
Output Partitioning of Neural Networks [PDF]
Many constructive learning algorithms have been proposed to find an appropriate network structure for a classification problem automatically. Constructive learning algorithms have drawbacks especially when used for complex tasks and modular approaches ...
Yinan, Q. +11 more
core +1 more source
Stochastic Control for Bayesian Neural Network Training
In this paper, we propose to leverage the Bayesian uncertainty information encoded in parameter distributions to inform the learning procedure for Bayesian models.
Ludwig Winkler +2 more
doaj +1 more source
Restricted Bayesian Neural Network
Modern deep learning tools are remarkably effective in addressing intricate problems. However, their operation as black-box models introduces increased uncertainty in predictions. Additionally, they contend with various challenges, including the need for substantial storage space in large networks, issues of overfitting, underfitting, vanishing ...
Ganguly, Sourav +1 more
openaire +3 more sources
To investigate and develop methods for the fitting of Bayesian neural ...
AlMalawi, Maysoon
core
Efficient Scaling of Bayesian Neural Networks
While Bayesian neural networks (BNNs) have gained popularity for their theoretical guarantees and robustness, they have yet to see a convincing implementation at scale.
Jacob R. Epifano +3 more
doaj +1 more source
Bayesian Sheaf Neural Networks
32 pages, 4 ...
Gillespie, Patrick +4 more
openaire +3 more sources
Bayesian Learning of Neural Network Architectures
The 22nd International Conference on Artificial Intelligence and Statistics (AISTATS 2019)
Georgi Dikov +2 more
openaire +2 more sources

