Dimensionality Reduction and Prior Knowledge in E-Set Recognition

By Kevin J. Lang, Geoffrey E. Hinton. 1989

Subfields: Representation & Self-Supervised Learning

How researchers see this paper

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

In Representation & Self-Supervised Learning

1. Nando de Freitas
2. Daniel Cremers
3. Marc'Aurelio Ranzato
4. Misha Denil
5. Vladimir Golkov
6. Babak Shakibi
7. Laurent Dinh
8. Jan Kukacka

Citations in the corpus

• Predicting Parameters in Deep Learning (2013, background): “Lang and Hinton [14] tried imposing explicit smoothness constraints through regularization but found it to universally reduce performance.”
• Regularization for Deep Learning: A Taxonomy (2017, background): “A classical regularizer is weight decay (see Plaut et al., 1986; Lang and Hinton, 1990; Goodfellow et al., 2016, Chap.”
• Neural networks for pattern recognition (1995, unknown): “= w 2 This problem can be overcome by using a modified decay term of the form (Hanson and Pratt , 1989; Lang and Hinton, 1990; Weigend et al, 1990)”