Matches in SemOpenAlex for { <https://semopenalex.org/work/W3102238049> ?p ?o ?g. }
- W3102238049 abstract "We present a computational method for empirically characterizing the training loss level-sets of deep neural networks. Our method numerically constructs a path in parameter space that is constrained to a set with a fixed near-zero training loss. By measuring regularization functions and test loss at different points within this path, we examine how different points in the parameter space with the same fixed training loss compare in terms of generalization ability. We also compare this method for finding regularized points with the more typical method, that uses objective functions which are weighted sums of training loss and regularization terms. We apply dimensionality reduction to the traversed paths in order to visualize the loss level sets in a well-regularized region of parameter space. Our results provide new information about the loss landscape of deep neural networks, as well as a new strategy for reducing test loss." @default.
- W3102238049 created "2020-11-23" @default.
- W3102238049 creator A5051693199 @default.
- W3102238049 creator A5085686472 @default.
- W3102238049 date "2020-11-09" @default.
- W3102238049 modified "2023-09-27" @default.
- W3102238049 title "Numerical Exploration of Training Loss Level-Sets in Deep Neural Networks" @default.
- W3102238049 cites W1481566577 @default.
- W3102238049 cites W1533861849 @default.
- W3102238049 cites W2001619934 @default.
- W3102238049 cites W2086513155 @default.
- W3102238049 cites W2095705004 @default.
- W3102238049 cites W2111072639 @default.
- W3102238049 cites W2112796928 @default.
- W3102238049 cites W2114766824 @default.
- W3102238049 cites W2118706537 @default.
- W3102238049 cites W2143101939 @default.
- W3102238049 cites W2296319761 @default.
- W3102238049 cites W2399994860 @default.
- W3102238049 cites W2401231614 @default.
- W3102238049 cites W2423689290 @default.
- W3102238049 cites W2788838181 @default.
- W3102238049 cites W2793333878 @default.
- W3102238049 cites W2945552729 @default.
- W3102238049 cites W2962715412 @default.
- W3102238049 cites W2962754331 @default.
- W3102238049 cites W2963008903 @default.
- W3102238049 cites W2963173418 @default.
- W3102238049 cites W2964121744 @default.
- W3102238049 cites W3118608800 @default.
- W3102238049 cites W61948969 @default.
- W3102238049 cites W639309485 @default.
- W3102238049 hasPublicationYear "2020" @default.
- W3102238049 type Work @default.
- W3102238049 sameAs 3102238049 @default.
- W3102238049 citedByCount "0" @default.
- W3102238049 crossrefType "posted-content" @default.
- W3102238049 hasAuthorship W3102238049A5051693199 @default.
- W3102238049 hasAuthorship W3102238049A5085686472 @default.
- W3102238049 hasConcept C111030470 @default.
- W3102238049 hasConcept C11413529 @default.
- W3102238049 hasConcept C126255220 @default.
- W3102238049 hasConcept C134306372 @default.
- W3102238049 hasConcept C154945302 @default.
- W3102238049 hasConcept C169903167 @default.
- W3102238049 hasConcept C177148314 @default.
- W3102238049 hasConcept C194273485 @default.
- W3102238049 hasConcept C199360897 @default.
- W3102238049 hasConcept C2776135515 @default.
- W3102238049 hasConcept C2777735758 @default.
- W3102238049 hasConcept C28826006 @default.
- W3102238049 hasConcept C2984842247 @default.
- W3102238049 hasConcept C33923547 @default.
- W3102238049 hasConcept C41008148 @default.
- W3102238049 hasConcept C50644808 @default.
- W3102238049 hasConcept C51632099 @default.
- W3102238049 hasConcept C555944384 @default.
- W3102238049 hasConcept C76155785 @default.
- W3102238049 hasConceptScore W3102238049C111030470 @default.
- W3102238049 hasConceptScore W3102238049C11413529 @default.
- W3102238049 hasConceptScore W3102238049C126255220 @default.
- W3102238049 hasConceptScore W3102238049C134306372 @default.
- W3102238049 hasConceptScore W3102238049C154945302 @default.
- W3102238049 hasConceptScore W3102238049C169903167 @default.
- W3102238049 hasConceptScore W3102238049C177148314 @default.
- W3102238049 hasConceptScore W3102238049C194273485 @default.
- W3102238049 hasConceptScore W3102238049C199360897 @default.
- W3102238049 hasConceptScore W3102238049C2776135515 @default.
- W3102238049 hasConceptScore W3102238049C2777735758 @default.
- W3102238049 hasConceptScore W3102238049C28826006 @default.
- W3102238049 hasConceptScore W3102238049C2984842247 @default.
- W3102238049 hasConceptScore W3102238049C33923547 @default.
- W3102238049 hasConceptScore W3102238049C41008148 @default.
- W3102238049 hasConceptScore W3102238049C50644808 @default.
- W3102238049 hasConceptScore W3102238049C51632099 @default.
- W3102238049 hasConceptScore W3102238049C555944384 @default.
- W3102238049 hasConceptScore W3102238049C76155785 @default.
- W3102238049 hasLocation W31022380491 @default.
- W3102238049 hasOpenAccess W3102238049 @default.
- W3102238049 hasPrimaryLocation W31022380491 @default.
- W3102238049 hasRelatedWork W1839868949 @default.
- W3102238049 hasRelatedWork W2496958009 @default.
- W3102238049 hasRelatedWork W2808242899 @default.
- W3102238049 hasRelatedWork W2910303224 @default.
- W3102238049 hasRelatedWork W2947664235 @default.
- W3102238049 hasRelatedWork W2963455631 @default.
- W3102238049 hasRelatedWork W2963982496 @default.
- W3102238049 hasRelatedWork W2964313743 @default.
- W3102238049 hasRelatedWork W2966530573 @default.
- W3102238049 hasRelatedWork W2968898205 @default.
- W3102238049 hasRelatedWork W2981748719 @default.
- W3102238049 hasRelatedWork W3046864714 @default.
- W3102238049 hasRelatedWork W3095344139 @default.
- W3102238049 hasRelatedWork W3106444498 @default.
- W3102238049 hasRelatedWork W3127478689 @default.
- W3102238049 hasRelatedWork W3153348027 @default.
- W3102238049 hasRelatedWork W3166144848 @default.
- W3102238049 hasRelatedWork W3200346102 @default.
- W3102238049 hasRelatedWork W99941565 @default.
- W3102238049 hasRelatedWork W2907171849 @default.