Insights into the mechanism of hydrate nucleation are of great significance for the development of hydrate-based technologies,hydrate relevant flow assurance,and the exploration of in situ natural gas hydrates.Compare...Insights into the mechanism of hydrate nucleation are of great significance for the development of hydrate-based technologies,hydrate relevant flow assurance,and the exploration of in situ natural gas hydrates.Compared with the thermodynamics of hydrate formation,understanding the nucleation mechanism is challenging and has drawn substantial attention in recent decades.In this paper,we attempt to give a comprehensive review of the recent progress of studies of clathrate hydrate nucleation.First,the existing hypotheses on the hydrate nucleation mechanism are introduced and discussed.Then,we summarize recent experimental studies on induction time,a key parameter evaluating the velocity of the nucleation process.Subsequently,the memory effect is particularly discussed,followed by the suggestion of several promising research perspectives.展开更多
Regarding extreme value theory,the unseen novelclasses in the openset recognition can be seen as the extremevalues of training classes.Following this idea,we introducethe margin and coverage distribution to model the ...Regarding extreme value theory,the unseen novelclasses in the openset recognition can be seen as the extremevalues of training classes.Following this idea,we introducethe margin and coverage distribution to model the trainingclasses.A novel visual-semantic embedding framework-extreme vocabulary learning(EVoL)is proposed;the EVoL embeds the visual features into semantic space in a probabilisticway.Notably,we adopt the vast open vocabulary in the semantic space to help further constraint the margin and coverage of training classes.The learned embedding can directlybe used to solve supervised learning,zero-shot learning,andopen set recognition simultaneously.Experiments on twobenchmark datasets demonstrate the effectiveness of the proposed framework against conventional ways.展开更多
基金Supported by the National Key Research and Development Program of China(2016YFC0304003,2017YFC0307302)the National Natural Science Foundation of China(21636009,51576209,21522609)the Innovative Talents Support Program for Postgraduates(BX201700288)
文摘Insights into the mechanism of hydrate nucleation are of great significance for the development of hydrate-based technologies,hydrate relevant flow assurance,and the exploration of in situ natural gas hydrates.Compared with the thermodynamics of hydrate formation,understanding the nucleation mechanism is challenging and has drawn substantial attention in recent decades.In this paper,we attempt to give a comprehensive review of the recent progress of studies of clathrate hydrate nucleation.First,the existing hypotheses on the hydrate nucleation mechanism are introduced and discussed.Then,we summarize recent experimental studies on induction time,a key parameter evaluating the velocity of the nucleation process.Subsequently,the memory effect is particularly discussed,followed by the suggestion of several promising research perspectives.
文摘Regarding extreme value theory,the unseen novelclasses in the openset recognition can be seen as the extremevalues of training classes.Following this idea,we introducethe margin and coverage distribution to model the trainingclasses.A novel visual-semantic embedding framework-extreme vocabulary learning(EVoL)is proposed;the EVoL embeds the visual features into semantic space in a probabilisticway.Notably,we adopt the vast open vocabulary in the semantic space to help further constraint the margin and coverage of training classes.The learned embedding can directlybe used to solve supervised learning,zero-shot learning,andopen set recognition simultaneously.Experiments on twobenchmark datasets demonstrate the effectiveness of the proposed framework against conventional ways.
