2924 search results - page 277 / 585 » On the Use of Evidence in Neural Networks |
174
click to vote
ML
1998
ACM
15 years 6 months ago
1998
ACM
Following Tesauro’s work on TD-Gammon, we used a 4000 parameter feed-forward neural network to develop a competitive backgammon evaluation function. Play proceeds by a roll of t...
170
click to vote
SENSYS
2006
ACM
16 years 19 days ago
2006
ACM
Small-form-factor, low-power wireless sensors—motes—are convenient to deploy, but lack the bandwidth to capture and transmit raw high-frequency data, such as human voices or n...
147
click to vote
NIPS
2001
15 years 8 months ago
2001
We present a new approach to bounding the true error rate of a continuous valued classifier based upon PAC-Bayes bounds. The method first constructs a distribution over classifier...
193
click to vote
GLVLSI
1998
IEEE
15 years 11 months ago
1998
IEEE
In this research, we devised a new simple technique for statically holding analog weights, which does not require periodic refreshing. It further contains a mechanism to locally u...
213
click to vote
NECO
2010
15 years 5 months ago
2010
Abstract: Reservoir Computing (RC) systems are powerful models for online computations on input sequences. They consist of a memoryless readout neuron which is trained on top of a ...