FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2308558753> ?p ?o ?g. }

• W2308558753 abstract "This thesis presents a clear conceptual basis for theoretically studying machine learning problems. Machine learning methods afford means to automate the discovery of relationships in data sets. A relationship between quantities X and Y allows the prediction of one quantity given information of the other. It is these relationships that we make the central object of study. We call these relationships transitions. A transition from a set X to a set Y is a function from X into the probability distributions on Y. Beginning with this simple notion, the thesis proceeds as follows: • Utilizing tools from statistical decision theory, we develop an abstract language for quantifying the information present in a transition. • We attack the problem of generalized supervision. Generalized supervision is the learning of classifiers from non-ideal data. An important example of this is the learning of classifiers from noisily labelled data. We demonstrate the virtues of our abstract treatment by producing generic methods for solving these problems, as well as producing generic upper bounds for our methods as well as lower bounds for any method that attempts to solve these problems. • As a result of our study in generalized supervision, we produce means to define procedures that are robust to certain forms of corruption. We explore, in detail, procedures for learning classifiers that are robust to the effects of symmetric label noise. The result is a classification algorithm that is easier to understand, implement and parallelize than standard kernel based classification schemes, such as the support vector machine and logistic regression. Furthermore, we demonstrate the uniqueness of this method. • Finally, we show how many feature learning schemes can be understood via our language. We present well motivated objectives for the task of learning features from unlabelled data, before showing how many standard feature learning methods (such as PCA, sparse coding, auto-encoders and so on) can be seen as minimizing surrogates to our objective functions." @default.
• W2308558753 created "2016-06-24" @default.
• W2308558753 creator A5050244538 @default.
• W2308558753 date "2015-09-01" @default.
• W2308558753 modified "2023-09-27" @default.
• W2308558753 title "Machine learning via transitions" @default.
• W2308558753 cites W1479807131 @default.
• W2308558753 cites W1501500081 @default.
• W2308558753 cites W1505771370 @default.
• W2308558753 cites W152055444 @default.
• W2308558753 cites W1524622012 @default.
• W2308558753 cites W1534582174 @default.
• W2308558753 cites W1535258871 @default.
• W2308558753 cites W1540637571 @default.
• W2308558753 cites W1551944942 @default.
• W2308558753 cites W1560724230 @default.
• W2308558753 cites W1564947197 @default.
• W2308558753 cites W1570963478 @default.
• W2308558753 cites W1573195610 @default.
• W2308558753 cites W1575388622 @default.
• W2308558753 cites W1582318892 @default.
• W2308558753 cites W1590693676 @default.
• W2308558753 cites W1607038179 @default.
• W2308558753 cites W1622633983 @default.
• W2308558753 cites W1666623353 @default.
• W2308558753 cites W1859987214 @default.
• W2308558753 cites W1866935739 @default.
• W2308558753 cites W1873158149 @default.
• W2308558753 cites W1873369706 @default.
• W2308558753 cites W1907483012 @default.
• W2308558753 cites W1946137962 @default.
• W2308558753 cites W1969244384 @default.
• W2308558753 cites W1975128126 @default.
• W2308558753 cites W1982032418 @default.
• W2308558753 cites W1990393033 @default.
• W2308558753 cites W1994046901 @default.
• W2308558753 cites W1994908596 @default.
• W2308558753 cites W1995897489 @default.
• W2308558753 cites W1997865285 @default.
• W2308558753 cites W2000015806 @default.
• W2308558753 cites W2004158585 @default.
• W2308558753 cites W2006258746 @default.
• W2308558753 cites W2014384147 @default.
• W2308558753 cites W2025720061 @default.
• W2308558753 cites W2025768430 @default.
• W2308558753 cites W2033468335 @default.
• W2308558753 cites W2041913211 @default.
• W2308558753 cites W2044828368 @default.
• W2308558753 cites W2046495522 @default.
• W2308558753 cites W2052100548 @default.
• W2308558753 cites W2053801139 @default.
• W2308558753 cites W2059944969 @default.
• W2308558753 cites W2060092029 @default.
• W2308558753 cites W2060322444 @default.
• W2308558753 cites W2068221105 @default.
• W2308558753 cites W2069735839 @default.
• W2308558753 cites W2070945723 @default.
• W2308558753 cites W2072403451 @default.
• W2308558753 cites W2073241381 @default.
• W2308558753 cites W2074950806 @default.
• W2308558753 cites W2087273767 @default.
• W2308558753 cites W2096520669 @default.
• W2308558753 cites W2096765209 @default.
• W2308558753 cites W2096840748 @default.
• W2308558753 cites W2099579348 @default.
• W2308558753 cites W2100495367 @default.
• W2308558753 cites W2101557761 @default.
• W2308558753 cites W2102348129 @default.
• W2308558753 cites W2105463715 @default.
• W2308558753 cites W2105464873 @default.
• W2308558753 cites W2106458073 @default.
• W2308558753 cites W2106491486 @default.
• W2308558753 cites W2107189314 @default.
• W2308558753 cites W2108198948 @default.
• W2308558753 cites W2108871462 @default.
• W2308558753 cites W2109706083 @default.
• W2308558753 cites W2110758705 @default.
• W2308558753 cites W2111181349 @default.
• W2308558753 cites W2111296615 @default.
• W2308558753 cites W2113290770 @default.
• W2308558753 cites W2116314036 @default.
• W2308558753 cites W2116470445 @default.
• W2308558753 cites W2119821739 @default.
• W2308558753 cites W2123602290 @default.
• W2308558753 cites W2128385268 @default.
• W2308558753 cites W2129000925 @default.
• W2308558753 cites W2130797782 @default.
• W2308558753 cites W2132283655 @default.
• W2308558753 cites W2134842679 @default.
• W2308558753 cites W2135046866 @default.
• W2308558753 cites W2135346645 @default.
• W2308558753 cites W2138056719 @default.
• W2308558753 cites W2144846366 @default.
• W2308558753 cites W2145295623 @default.
• W2308558753 cites W2148603752 @default.
• W2308558753 cites W2149175418 @default.
• W2308558753 cites W2149570649 @default.
• W2308558753 cites W2153190668 @default.
• W2308558753 cites W2154318594 @default.
• W2308558753 cites W2155319834 @default.

• next