Neural Networks: A Comprehensive Foundation

By Simon S. Haykin. 1998

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Opening the Black Box of Deep Neural Networks via Information (2017, methodology): “answer the ﬁrst question, we repeated our information plane and stochastic gradient analysis with an entirely different non-symmetric rule and architectures - the well studied committee machine [e.g. Haykin (1998)]. As can be seen in ﬁgure 8 the Information Plane paths and the SGD phases are very si”
Breaking the Curse of Dimensionality with Convex Neural Networks (2014, background): “the 2d −1 non-trivial potential subsets. (d) Single hidden-layer neural networks: f(x) = Pk j=1 σ(w ⊤ j x + bj), where k is the number of units in the hidden layer (see, e.g., Rumelhart et al., 1986; Haykin, 1994). The activation function σ is here assumed to be ﬁxed. While the learning problem may ”
A statistical approach to rule learning (2006, methodology): “For example, there are algorithms for inducing neural networks (Haykin, 1999): in this representation, a number of artificial neurons are connected with weighted connections, often in a hierarchical structure.”
Support Vector Machines Used to Estimate the Battery State of Charge (2013, methodology): “Like conventional feedforward neural networks, such as the multilayer perceptron [19], SVM has been used by researchers to solve classification and regression problems [20].”
Recognizing human motion with multiple acceleration sensors (2001, methodology): “PCA and ICA are well-established statistical tools, in the signal processing and data-analysis communities [ 5 ], [ll].”
Selecting appropriate machine learning methods for digital soil mapping (2020, background): “(speed of convergence for the gradient descent learning algorithm) [49].”