FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3005871659> ?p ?o ?g. }

• W3005871659 abstract "Robust loss minimization is an important strategy for handling robust learning issue on noisy labels. Current robust loss functions, however, inevitably involve hyperparameter(s) to be tuned, manually or heuristically through cross validation, which makes them fairly hard to be generally applied in practice. Besides, the non-convexity brought by the loss as well as the complicated network architecture makes it easily trapped into an unexpected solution with poor generalization capability. To address above issues, we propose a meta-learning method capable of adaptively learning hyperparameter in robust loss functions. Specifically, through mutual amelioration between robust loss hyperparameter and network parameters in our method, both of them can be simultaneously finely learned and coordinated to attain solutions with good generalization capability. Four kinds of SOTA robust loss functions are attempted to be integrated into our algorithm, and comprehensive experiments substantiate the general availability and effectiveness of the proposed method in both its accuracy and generalization performance, as compared with conventional hyperparameter tuning strategy, even with carefully tuned hyperparameters." @default.
• W3005871659 created "2020-02-24" @default.
• W3005871659 creator A5004966418 @default.
• W3005871659 creator A5075938156 @default.
• W3005871659 creator A5078092645 @default.
• W3005871659 creator A5081768187 @default.
• W3005871659 creator A5091017287 @default.
• W3005871659 date "2020-02-16" @default.
• W3005871659 modified "2023-09-27" @default.
• W3005871659 title "Learning Adaptive Loss for Robust Learning with Noisy Labels." @default.
• W3005871659 cites W1921293667 @default.
• W3005871659 cites W1975128126 @default.
• W3005871659 cites W1982032418 @default.
• W3005871659 cites W1995137594 @default.
• W3005871659 cites W2032708784 @default.
• W3005871659 cites W2089217417 @default.
• W3005871659 cites W2097998348 @default.
• W3005871659 cites W2102348129 @default.
• W3005871659 cites W2108018401 @default.
• W3005871659 cites W2113290770 @default.
• W3005871659 cites W2131241448 @default.
• W3005871659 cites W2136688338 @default.
• W3005871659 cites W2155964055 @default.
• W3005871659 cites W2163605009 @default.
• W3005871659 cites W2187089797 @default.
• W3005871659 cites W2194775991 @default.
• W3005871659 cites W2394968026 @default.
• W3005871659 cites W2557912764 @default.
• W3005871659 cites W2578908617 @default.
• W3005871659 cites W2592335154 @default.
• W3005871659 cites W2620239873 @default.
• W3005871659 cites W2752971446 @default.
• W3005871659 cites W2810089099 @default.
• W3005871659 cites W2886219088 @default.
• W3005871659 cites W2899771611 @default.
• W3005871659 cites W2945634539 @default.
• W3005871659 cites W2946154049 @default.
• W3005871659 cites W2946707734 @default.
• W3005871659 cites W2962762068 @default.
• W3005871659 cites W2962762541 @default.
• W3005871659 cites W2962826047 @default.
• W3005871659 cites W2963081269 @default.
• W3005871659 cites W2963233958 @default.
• W3005871659 cites W2963306862 @default.
• W3005871659 cites W2963371670 @default.
• W3005871659 cites W2963476860 @default.
• W3005871659 cites W2963697299 @default.
• W3005871659 cites W2963759070 @default.
• W3005871659 cites W2963780286 @default.
• W3005871659 cites W2963803100 @default.
• W3005871659 cites W2963804140 @default.
• W3005871659 cites W2963826056 @default.
• W3005871659 cites W2964096266 @default.
• W3005871659 cites W2964274690 @default.
• W3005871659 cites W2964292098 @default.
• W3005871659 cites W2964309657 @default.
• W3005871659 cites W2970308742 @default.
• W3005871659 cites W2970711760 @default.
• W3005871659 cites W2971364389 @default.
• W3005871659 cites W2973562770 @default.
• W3005871659 cites W2981873476 @default.
• W3005871659 cites W3118608800 @default.
• W3005871659 cites W3137695714 @default.
• W3005871659 cites W369786348 @default.
• W3005871659 cites W880548201 @default.
• W3005871659 cites W99485931 @default.
• W3005871659 hasPublicationYear "2020" @default.
• W3005871659 type Work @default.
• W3005871659 sameAs 3005871659 @default.
• W3005871659 citedByCount "9" @default.
• W3005871659 countsByYear W30058716592020 @default.
• W3005871659 countsByYear W30058716592021 @default.
• W3005871659 crossrefType "posted-content" @default.
• W3005871659 hasAuthorship W3005871659A5004966418 @default.
• W3005871659 hasAuthorship W3005871659A5075938156 @default.
• W3005871659 hasAuthorship W3005871659A5078092645 @default.
• W3005871659 hasAuthorship W3005871659A5081768187 @default.
• W3005871659 hasAuthorship W3005871659A5091017287 @default.
• W3005871659 hasConcept C119857082 @default.
• W3005871659 hasConcept C134306372 @default.
• W3005871659 hasConcept C154945302 @default.
• W3005871659 hasConcept C177148314 @default.
• W3005871659 hasConcept C33923547 @default.
• W3005871659 hasConcept C41008148 @default.
• W3005871659 hasConcept C8642999 @default.
• W3005871659 hasConceptScore W3005871659C119857082 @default.
• W3005871659 hasConceptScore W3005871659C134306372 @default.
• W3005871659 hasConceptScore W3005871659C154945302 @default.
• W3005871659 hasConceptScore W3005871659C177148314 @default.
• W3005871659 hasConceptScore W3005871659C33923547 @default.
• W3005871659 hasConceptScore W3005871659C41008148 @default.
• W3005871659 hasConceptScore W3005871659C8642999 @default.
• W3005871659 hasLocation W30058716591 @default.
• W3005871659 hasOpenAccess W3005871659 @default.
• W3005871659 hasPrimaryLocation W30058716591 @default.
• W3005871659 hasRelatedWork W1514928307 @default.
• W3005871659 hasRelatedWork W1921293667 @default.
• W3005871659 hasRelatedWork W2113290770 @default.
• W3005871659 hasRelatedWork W2810089099 @default.
• W3005871659 hasRelatedWork W2945634539 @default.

• next