FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2187000089> ?p ?o ?g. }

• W2187000089 abstract "Virtually all methods of learning dynamic models from data start from the same basic assumption: that the learning algorithm will be provided with a single or multiple sequences of data generated from the dynamic model. In this work we consider the case where the data is not sequenced. The learning algorithm is presented a set of data points from the system’s operation but with no temporal ordering. The data are simply drawn as individual disconnected points, and usually come from many separate executions of the dynamic system. Many scienti c modeling tasks have exactly this property. Consider the task of learning dynamic models of galaxy or star evolution. The dynamics of these processes are far too slow for us to collect successive data points showing any meaningful changes. However, we do have billions of single data points showing these objects at various stages of their evolution. At the other end of the spectrum, cellular or molecular biological processes may be too small or too fast to permit collection of trajectories from the system. Often, the measurement techniques are destructive and thus only one data point can be collected from each sample even though a rough indication of the relative timing between samples may be known. In all of these applications, scientists would like to construct a dynamic model using only nonsequenced, individual data points collected from the system of interest. In this work, we launch investigation of this topic by considering fully observable, continuous-state, discrete-time models. We formalize the task of learning from non-sequenced data under linear models, and discuss several issues in identi ability. We then develop several learning algorithms based on optimizing approximate posterior distributions, and generalize these algorithms to learn nonlinear models through reproducing kernels. These proposed methods essentially carry out expectation maximization, and thus require good initialization. We address this issue by studying a related problem, that of reconstructing a temporal sequence from out-of-order data points. To solve this problem, we propose a convex program that optimizes a measure of temporal smoothness, and develop e cient solution algorithms. We test these methods on several synthetic and real data sets, including a set of gene expression time series data collected from yeast. Experimental results show both e ectiveness and limitations of the proposed methods." @default.
• W2187000089 created "2016-06-24" @default.
• W2187000089 creator A5026244838 @default.
• W2187000089 creator A5074586221 @default.
• W2187000089 date "2010-01-01" @default.
• W2187000089 modified "2023-09-24" @default.
• W2187000089 title "Learning Dynamic Models from Non-sequenced Data" @default.
• W2187000089 cites W1536329667 @default.
• W2187000089 cites W1566076669 @default.
• W2187000089 cites W1574662932 @default.
• W2187000089 cites W1576002161 @default.
• W2187000089 cites W1578316706 @default.
• W2187000089 cites W164888260 @default.
• W2187000089 cites W1663973292 @default.
• W2187000089 cites W1960700909 @default.
• W2187000089 cites W1965324089 @default.
• W2187000089 cites W1972316051 @default.
• W2187000089 cites W1978168102 @default.
• W2187000089 cites W1978259121 @default.
• W2187000089 cites W1988615825 @default.
• W2187000089 cites W2006912660 @default.
• W2187000089 cites W2017588182 @default.
• W2187000089 cites W2036799174 @default.
• W2187000089 cites W2088563154 @default.
• W2187000089 cites W2096389124 @default.
• W2187000089 cites W2097150846 @default.
• W2187000089 cites W2098460974 @default.
• W2187000089 cites W2102716594 @default.
• W2187000089 cites W2108414494 @default.
• W2187000089 cites W2108688899 @default.
• W2187000089 cites W2110575115 @default.
• W2187000089 cites W2113256684 @default.
• W2187000089 cites W2125838338 @default.
• W2187000089 cites W2138153286 @default.
• W2187000089 cites W2139302369 @default.
• W2187000089 cites W2141394518 @default.
• W2187000089 cites W2158190429 @default.
• W2187000089 cites W2168834831 @default.
• W2187000089 cites W2569025259 @default.
• W2187000089 cites W2799004609 @default.
• W2187000089 cites W1541680420 @default.
• W2187000089 hasPublicationYear "2010" @default.
• W2187000089 type Work @default.
• W2187000089 sameAs 2187000089 @default.
• W2187000089 citedByCount "0" @default.
• W2187000089 crossrefType "journal-article" @default.
• W2187000089 hasAuthorship W2187000089A5026244838 @default.
• W2187000089 hasAuthorship W2187000089A5074586221 @default.
• W2187000089 hasConcept C111472728 @default.
• W2187000089 hasConcept C11413529 @default.
• W2187000089 hasConcept C119857082 @default.
• W2187000089 hasConcept C121332964 @default.
• W2187000089 hasConcept C124101348 @default.
• W2187000089 hasConcept C138885662 @default.
• W2187000089 hasConcept C154945302 @default.
• W2187000089 hasConcept C162324750 @default.
• W2187000089 hasConcept C177264268 @default.
• W2187000089 hasConcept C187736073 @default.
• W2187000089 hasConcept C189950617 @default.
• W2187000089 hasConcept C197298091 @default.
• W2187000089 hasConcept C199360897 @default.
• W2187000089 hasConcept C2524010 @default.
• W2187000089 hasConcept C2780451532 @default.
• W2187000089 hasConcept C2780801425 @default.
• W2187000089 hasConcept C28719098 @default.
• W2187000089 hasConcept C32848918 @default.
• W2187000089 hasConcept C33923547 @default.
• W2187000089 hasConcept C41008148 @default.
• W2187000089 hasConcept C58489278 @default.
• W2187000089 hasConcept C62520636 @default.
• W2187000089 hasConcept C80444323 @default.
• W2187000089 hasConceptScore W2187000089C111472728 @default.
• W2187000089 hasConceptScore W2187000089C11413529 @default.
• W2187000089 hasConceptScore W2187000089C119857082 @default.
• W2187000089 hasConceptScore W2187000089C121332964 @default.
• W2187000089 hasConceptScore W2187000089C124101348 @default.
• W2187000089 hasConceptScore W2187000089C138885662 @default.
• W2187000089 hasConceptScore W2187000089C154945302 @default.
• W2187000089 hasConceptScore W2187000089C162324750 @default.
• W2187000089 hasConceptScore W2187000089C177264268 @default.
• W2187000089 hasConceptScore W2187000089C187736073 @default.
• W2187000089 hasConceptScore W2187000089C189950617 @default.
• W2187000089 hasConceptScore W2187000089C197298091 @default.
• W2187000089 hasConceptScore W2187000089C199360897 @default.
• W2187000089 hasConceptScore W2187000089C2524010 @default.
• W2187000089 hasConceptScore W2187000089C2780451532 @default.
• W2187000089 hasConceptScore W2187000089C2780801425 @default.
• W2187000089 hasConceptScore W2187000089C28719098 @default.
• W2187000089 hasConceptScore W2187000089C32848918 @default.
• W2187000089 hasConceptScore W2187000089C33923547 @default.
• W2187000089 hasConceptScore W2187000089C41008148 @default.
• W2187000089 hasConceptScore W2187000089C58489278 @default.
• W2187000089 hasConceptScore W2187000089C62520636 @default.
• W2187000089 hasConceptScore W2187000089C80444323 @default.
• W2187000089 hasLocation W21870000891 @default.
• W2187000089 hasOpenAccess W2187000089 @default.
• W2187000089 hasPrimaryLocation W21870000891 @default.
• W2187000089 hasRelatedWork W1494236600 @default.
• W2187000089 hasRelatedWork W1531643234 @default.
• W2187000089 hasRelatedWork W1554288356 @default.

• next