Sciweavers

107 search results - page 9 / 22
» Quantum Circuits: From a Network to a One-Way Model
Sort
View
FLAIRS
2006
13 years 9 months ago
Decomposing Local Probability Distributions in Bayesian Networks for Improved Inference and Parameter Learning
A major difficulty in building Bayesian network models is the size of conditional probability tables, which grow exponentially in the number of parents. One way of dealing with th...
Adam Zagorecki, Mark Voortman, Marek J. Druzdzel
TNN
1998
114views more  TNN 1998»
13 years 7 months ago
A new approach to artificial neural networks
: A novel approach to artificial neural networks is presented. The philosophy of this approach is based on two aspects: the design of task-specific networks, and a new neuron model...
Benedito Dias Baptista F. Filho, Eduardo Lobo Lust...
TIT
2008
83views more  TIT 2008»
13 years 7 months ago
Constrained Codes as Networks of Relations
We address the well-known problem of determining the capacity of constrained coding systems. While the onedimensional case is well understood to the extent that there are technique...
Moshe Schwartz, Jehoshua Bruck
IJCNN
2000
IEEE
13 years 12 months ago
A 2D Neuromorphic VLSI Architecture for Modeling Selective Attention
Selectiveattentionis a mechanismsused to sequentiallyselectthe spatiallocationsof salientregionsin the sensor’sfieldof view. This mechanism overcomesthe problem of flooding limi...
Giacomo Indiveri
COLT
2006
Springer
13 years 11 months ago
Efficient Learning Algorithms Yield Circuit Lower Bounds
We describe a new approach for understanding the difficulty of designing efficient learning algorithms. We prove that the existence of an efficient learning algorithm for a circui...
Lance Fortnow, Adam R. Klivans