The Tempo 2 Algorithm: Adjusting Time-Delays By Supervised Learning

By Ulrich Bodenhausen, A. Waibel. 1990

How researchers see this paper

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

1. J. Urgen Schmidhuber
2. Mark B. Ring
3. Yoshua Bengio
4. Michael C. Mozer
5. Sepp Hochreiter
6. Sergey Levine
7. Quoc V. Le
8. Trevor Darrell

Citations in the corpus

• Deep learning in neural networks: An overview (2014, methodology): “s were proposed, including Focused BP based on decay factors for activations of units in RNNs (Mozer, 1989, 1992), Time-Delay Neural Networks (TDNNs) (Lang et al., 1990) and their adaptive extension (Bodenhausen and Waibel, 1991), Nonlinear AutoRegressive with eXogenous inputs (NARX) RNNs (Lin et al”
• Learning Unambiguous Reduced Sequence Descriptions (1991, methodology): “Instead, use your sequence learning algorithm (any will do) to implement the following method for history compression.”
• Continual learning in reinforcement environments (1995, background): “Another is Bodenhausen and Waibel's system, Tempo 2 [13].”
• Neural net architectures for temporal sequence processing (2007, background): “Interesting cases of adaptive memory models in the neural net literature include: the learning of delays (Bodenhausen & Waibel, 1991; Unnikrishnan, Hopfield, & Tank, 1991), the learning of exponential decay rates (Bachrach, 1988; Frasconi, Gori, & Soda, 1992; Mozer, 1989), and the learning of gamma…”