Matches in SemOpenAlex for { <https://semopenalex.org/work/W2894876375> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W2894876375 abstract "We propose a novel framework for structured prediction via adversarial learning. Existing adversarial learning methods involve two separate networks, i.e., the structured prediction models and the discriminative models, in the training. The information captured by discriminative models complements that in the structured prediction models, but few existing researches have studied on utilizing such information to improve structured prediction models at the inference stage. In this work, we propose to refine the predictions of structured prediction models by effectively integrating discriminative models into the prediction. Discriminative models are treated as energy-based models. Similar to the adversarial learning, discriminative models are trained to estimate scores which measure the quality of predicted outputs, while structured prediction models are trained to predict contrastive outputs with maximal energy scores. In this way, the gradient vanishing problem is ameliorated, and thus we are able to perform inference by following the ascent gradient directions of discriminative models to refine structured prediction models. The proposed method is able to handle a range of tasks, e.g., multi-label classification and image segmentation. Empirical results on these two tasks validate the effectiveness of our learning method." @default.
- W2894876375 created "2018-10-12" @default.
- W2894876375 creator A5013166957 @default.
- W2894876375 creator A5023288846 @default.
- W2894876375 creator A5048411184 @default.
- W2894876375 creator A5072690470 @default.
- W2894876375 date "2018-10-02" @default.
- W2894876375 modified "2023-09-23" @default.
- W2894876375 title "Learning Discriminators as Energy Networks in Adversarial Learning" @default.
- W2894876375 cites W1522301498 @default.
- W2894876375 cites W1548595933 @default.
- W2894876375 cites W1782590233 @default.
- W2894876375 cites W1903029394 @default.
- W2894876375 cites W1945616565 @default.
- W2894876375 cites W2099471712 @default.
- W2894876375 cites W2114096665 @default.
- W2894876375 cites W2124592697 @default.
- W2894876375 cites W2131984901 @default.
- W2894876375 cites W2158097779 @default.
- W2894876375 cites W2161914416 @default.
- W2894876375 cites W2226771013 @default.
- W2894876375 cites W2271840356 @default.
- W2894876375 cites W2521028896 @default.
- W2894876375 cites W2525954470 @default.
- W2894876375 cites W2555437177 @default.
- W2894876375 cites W2580360036 @default.
- W2894876375 cites W2581637843 @default.
- W2894876375 cites W2592480533 @default.
- W2894876375 cites W2593414223 @default.
- W2894876375 cites W2599275287 @default.
- W2894876375 cites W2604538595 @default.
- W2894876375 cites W2617781042 @default.
- W2894876375 cites W2766527293 @default.
- W2894876375 cites W2804461692 @default.
- W2894876375 cites W2950040358 @default.
- W2894876375 cites W2951140085 @default.
- W2894876375 cites W2952936466 @default.
- W2894876375 cites W2962793481 @default.
- W2894876375 cites W2962879692 @default.
- W2894876375 cites W2963073614 @default.
- W2894876375 cites W2963789586 @default.
- W2894876375 cites W2964205912 @default.
- W2894876375 cites W3140910462 @default.
- W2894876375 cites W648143168 @default.
- W2894876375 doi "https://doi.org/10.48550/arxiv.1810.01152" @default.
- W2894876375 hasPublicationYear "2018" @default.
- W2894876375 type Work @default.
- W2894876375 sameAs 2894876375 @default.
- W2894876375 citedByCount "2" @default.
- W2894876375 countsByYear W28948763752019 @default.
- W2894876375 crossrefType "posted-content" @default.
- W2894876375 hasAuthorship W2894876375A5013166957 @default.
- W2894876375 hasAuthorship W2894876375A5023288846 @default.
- W2894876375 hasAuthorship W2894876375A5048411184 @default.
- W2894876375 hasAuthorship W2894876375A5072690470 @default.
- W2894876375 hasBestOaLocation W28948763751 @default.
- W2894876375 hasConcept C108583219 @default.
- W2894876375 hasConcept C119857082 @default.
- W2894876375 hasConcept C153180895 @default.
- W2894876375 hasConcept C154945302 @default.
- W2894876375 hasConcept C22367795 @default.
- W2894876375 hasConcept C2776214188 @default.
- W2894876375 hasConcept C37736160 @default.
- W2894876375 hasConcept C41008148 @default.
- W2894876375 hasConcept C97931131 @default.
- W2894876375 hasConceptScore W2894876375C108583219 @default.
- W2894876375 hasConceptScore W2894876375C119857082 @default.
- W2894876375 hasConceptScore W2894876375C153180895 @default.
- W2894876375 hasConceptScore W2894876375C154945302 @default.
- W2894876375 hasConceptScore W2894876375C22367795 @default.
- W2894876375 hasConceptScore W2894876375C2776214188 @default.
- W2894876375 hasConceptScore W2894876375C37736160 @default.
- W2894876375 hasConceptScore W2894876375C41008148 @default.
- W2894876375 hasConceptScore W2894876375C97931131 @default.
- W2894876375 hasLocation W28948763751 @default.
- W2894876375 hasOpenAccess W2894876375 @default.
- W2894876375 hasPrimaryLocation W28948763751 @default.
- W2894876375 hasRelatedWork W1652783584 @default.
- W2894876375 hasRelatedWork W1781547478 @default.
- W2894876375 hasRelatedWork W2024160000 @default.
- W2894876375 hasRelatedWork W2353457699 @default.
- W2894876375 hasRelatedWork W2404514746 @default.
- W2894876375 hasRelatedWork W2511279186 @default.
- W2894876375 hasRelatedWork W2953238046 @default.
- W2894876375 hasRelatedWork W2963058055 @default.
- W2894876375 hasRelatedWork W4300631627 @default.
- W2894876375 hasRelatedWork W4319994054 @default.
- W2894876375 isParatext "false" @default.
- W2894876375 isRetracted "false" @default.
- W2894876375 magId "2894876375" @default.
- W2894876375 workType "article" @default.