intelligence had focused on high-level (symbolic) models that are processed by using algorithms, characterized for example by expert systems with knowledge embodied in if-then rules, until in the late 1980s research expanded to low-level (sub-symbolic) machine learning, characterized by knowledge embodied. Backpropagation training algorithms fall into three categories: Evolutionary methods, 88 gene expression programming, 89 simulated annealing, 90 expectation-maximization, non-parametric methods and particle swarm optimization 91 are other methods for training neural networks. "An artificial neural network approach to rainfall-runoff modelling". Kruse, Rudolf Borgelt, Christian; Klawonn,.; Moewes, Christian; Steinbrecher, Matthias; Held, Pascal (2013). Unsupervised learning edit In unsupervised learning, some data xdisplaystyle textstyle x is given and the cost function to be minimized, that can be any function of the data xdisplaystyle textstyle x and the network's output, fdisplaystyle textstyle. Eiji Mizutani, Stuart Dreyfus, Kenichi Nishio (2000).
Nanodevices 30 for very large scale principal components analyses and convolution may create a new class of neural computing because they are fundamentally analog rather than digital (even though the first implementations may use digital devices). AE is an unsupervised algorithm and it is different from the other models used in the paper since it learns the implicit distribution of the training data instead of mere discriminant features. An unreadable table that a useful machine could read would still be well worth having. 224 Capacity edit Models' "capacity" property roughly corresponds to their ability to model any given function. Alternatives to backpropagation include Extreme Learning Machines, 74 "No-prop" networks, 75 training without backtracking, 76 "weightless" networks, 77 78 and non-connectionist neural networks. 148 Tensor deep stacking networks edit This architecture is a DSN extension. A b Graves,.; Liwicki,.; Fernandez,.; Bertolami,.; Bunke,.; Schmidhuber,.
Research paper on neural networks