Results 271 to 280 of about 160,175 (304)
Demonstrating completeness in optical neural computing. [PDF]
Light Sci ApplTyszka K.
europepmc +1 more source
Local stability of Belief Propagation algorithm with multiple fixed points
, 2014 Victorin Martin +3 more
openalex +1 more source
Research on Joint Game-Theoretic Modeling of Network Attack and Defense Under Incomplete Information. [PDF]
Entropy (Basel)Wang Y, Liu X, Yu X.
europepmc +1 more source
Widely cited global irrigation statistics lack empirical support. [PDF]
PNAS NexusPuy A +8 more
europepmc +1 more source
The neuroscience of algorithmic suffering: short comparative analysis between human and AI. [PDF]
Front PsycholTütüncü EK, Gonzalez-Franco M.
europepmc +1 more source
Some of the next articles are maybe not open access.
Related searches:
Related searches:
Hardware-efficient belief propagation
2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009Loopy belief propagation (BP) is an effective solution for assigning labels to the nodes of a graphical model such as the Markov random field (MRF), but it requires high memory, bandwidth, and computational costs. Furthermore, the iterative, pixel-wise, and sequential operations of BP make it difficult to parallelize the computation.
null Chia-Kai Liang +4 more
openaire +1 more source
Motion Estimation via Belief Propagation
14th International Conference on Image Analysis and Processing (ICIAP 2007), 2007We present a probabilistic model for motion estimation in which motion characteristics are inferred on the basis of a finite mixture of motion models. The model is graphically represented in the form of a pairwise Markov Random Field network upon which a Loopy Belief Propagation algorithm is exploited to perform inference.
Giuseppe Boccignone +3 more
openaire +4 more sources