This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
publications [2020/07/09 04:53] admin |
publications [2020/07/09 04:55] admin |
||
---|---|---|---|
Line 3: | Line 3: | ||
=== Notable achievements: === | === Notable achievements: === | ||
- | * 1986: First implementation of Deep Neural Networks ([[https://www.researchgate.net/publication/338188874_Speech_Recognition_with_Associative_Networks|Speech Recognition with Associative Networks.]]) - these had the form o<sub>i</sub> = f(∑<sup>i-1</sup> w<sub>ij</sub> o<sub>j</sub>) and so were very deep having yet trained well using all possible skip connections. | + | * 1986: First implementation of Deep Neural Networks ([[https://www.researchgate.net/publication/338188874_Speech_Recognition_with_Associative_Networks|Speech Recognition with Associative Networks.]]) - these had the form x<sub>i</sub> = f(∑<sup>i-1</sup> w<sub>ij</sub> x<sub>j</sub>) and so were maximally deep yet trained well using all possible skip connections. |
* 1987: First publication of Real Time Recurrent Learning ([[https://www.academia.edu/30351853/The_utility_driven_dynamic_error_propagation_network|The utility driven dynamic error propagation network.]] also [[https://papers.nips.cc/paper/42-static-and-dynamic-error-propagation-networks-with-application-to-speech-coding|Static and dynamic error propagation networks with application to speech coding.]]) | * 1987: First publication of Real Time Recurrent Learning ([[https://www.academia.edu/30351853/The_utility_driven_dynamic_error_propagation_network|The utility driven dynamic error propagation network.]] also [[https://papers.nips.cc/paper/42-static-and-dynamic-error-propagation-networks-with-application-to-speech-coding|Static and dynamic error propagation networks with application to speech coding.]]) | ||
* 1991: First state-of-the-art ASR with neural networks ([[http://mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/robinson_tr82.pdf|Several improvements to a recurrent error propagation network phone recognition system]]) | * 1991: First state-of-the-art ASR with neural networks ([[http://mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/robinson_tr82.pdf|Several improvements to a recurrent error propagation network phone recognition system]]) |