@patent{yu_method_2019,
	title = {Method for {Meta}-{Level} {Continual} {Learning}},
	url = {https://patents.google.com/patent/US20190034798A1/en},
	abstract = {Classification of an input task data set by meta level continual learning includes analyzing first and second training data sets in a task space to generate first and second meta weights and a slow weight value, and comparing an input task data set to the slow weight to generate a fast weight. The first and second meta weights are parameterized with the fast weight value to update the slow weight value, whereby a value is associated with the input task data set, thereby classifying the input task data set by meta level continual learning.},
	nationality = {US},
	assignee = {University Of Massachusetts Medical School},
	number = {US20190034798A1},
	urldate = {2019-04-10},
	author = {Yu, Hong and Munkhdalai, Tsendsuren},
	month = jan,
	year = {2019},
	keywords = {loss, meta, slow, task, weight},
}

{"_id":"cb6d6a86WTR3nAYJn","bibbaseid":"yu-munkhdalai-methodformetalevelcontinuallearning-2019","author_short":["Yu, H.","Munkhdalai, T."],"bibdata":{"bibtype":"patent","type":"patent","title":"Method for Meta-Level Continual Learning","url":"https://patents.google.com/patent/US20190034798A1/en","abstract":"Classification of an input task data set by meta level continual learning includes analyzing first and second training data sets in a task space to generate first and second meta weights and a slow weight value, and comparing an input task data set to the slow weight to generate a fast weight. The first and second meta weights are parameterized with the fast weight value to update the slow weight value, whereby a value is associated with the input task data set, thereby classifying the input task data set by meta level continual learning.","nationality":"US","assignee":"University Of Massachusetts Medical School","number":"US20190034798A1","urldate":"2019-04-10","author":[{"propositions":[],"lastnames":["Yu"],"firstnames":["Hong"],"suffixes":[]},{"propositions":[],"lastnames":["Munkhdalai"],"firstnames":["Tsendsuren"],"suffixes":[]}],"month":"January","year":"2019","keywords":"loss, meta, slow, task, weight","bibtex":"@patent{yu_method_2019,\n\ttitle = {Method for {Meta}-{Level} {Continual} {Learning}},\n\turl = {https://patents.google.com/patent/US20190034798A1/en},\n\tabstract = {Classification of an input task data set by meta level continual learning includes analyzing first and second training data sets in a task space to generate first and second meta weights and a slow weight value, and comparing an input task data set to the slow weight to generate a fast weight. The first and second meta weights are parameterized with the fast weight value to update the slow weight value, whereby a value is associated with the input task data set, thereby classifying the input task data set by meta level continual learning.},\n\tnationality = {US},\n\tassignee = {University Of Massachusetts Medical School},\n\tnumber = {US20190034798A1},\n\turldate = {2019-04-10},\n\tauthor = {Yu, Hong and Munkhdalai, Tsendsuren},\n\tmonth = jan,\n\tyear = {2019},\n\tkeywords = {loss, meta, slow, task, weight},\n}\n\n","author_short":["Yu, H.","Munkhdalai, T."],"key":"yu_method_2019","id":"yu_method_2019","bibbaseid":"yu-munkhdalai-methodformetalevelcontinuallearning-2019","role":"author","urls":{"Paper":"https://patents.google.com/patent/US20190034798A1/en"},"keyword":["loss","meta","slow","task","weight"],"metadata":{"authorlinks":{}}},"bibtype":"patent","biburl":"http://fenway.cs.uml.edu/papers/pubs-all.bib","dataSources":["TqaA9miSB65nRfS5H"],"keywords":["loss","meta","slow","task","weight"],"search_terms":["method","meta","level","continual","learning","yu","munkhdalai"],"title":"Method for Meta-Level Continual Learning","year":2019}