Derrick Takeshi Mirikitani

Learn More
This paper develops a probabilistic approach to recursive second-order training of recurrent neural networks (RNNs) for improved time-series modeling. A general recursive Bayesian Levenberg-Marquardt algorithm is derived to sequentially update the weights and the covariance (Hessian) matrix. The main strengths of the approach are a principled handling of(More)
Although search engines are often used for information retrieval (IR) from the World Wide Web (WWW), current search engine technology seems obsolete. The quality of query results from today's search engines is unacceptable, creating a demand for new information search and retrieval techniques. The conventional IR methods often lack the flexibility to adapt(More)