Efficient Parallel Learning Algorithms for Neural Networks
Alan Kramer, Alberto Luigi Sangiovanni-Vincentelli · neural information processing systems · 1988
Parallelizable optimization techniques are applied to the problem of learning in feedforward neural networks. In addition to having superior convergence properties, optimization techniques such as the Polak-Ribiere method are also significantly more efficient than the Backpropagation algorithm. These results are based on experiments performed on small boolean learning problems and the noisy real-valued learning problem of hand-written character recognition.
Cited by 132 works.
Authors
References
- Learning Internal Representations by Error Propagation
- Linear and nonlinear programming.
- Improving the convergence of back-propagation learning with second-order methods
- The Connection Machine
- Learning distributed representations of concepts.
- Restart procedures for the conjugate gradient method
- Fast-learning variations on back propagation: an empirical study.
- Nonlinear Programming: Analysis and Methods
- The Connection Machine