Matches in SemOpenAlex for { <https://semopenalex.org/work/W3132000425> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W3132000425 abstract "In this work we introduce a simple baseline for meta-learning. Our unconventional method, FIX-ML, reduces task diversity by keeping support sets fixed across tasks, and consistently improves the performance of meta-learning methods on popular few-shot learning benchmarks. However, in exploring the reason for this counter-intuitive phenomenon, we unearth a series of questions and concerns about meta-learning evaluation practices. We explore two possible goals of meta-learning: to develop methods that generalize (i) to the same task distribution that generates the training set (in-distribution), or (ii) to new, unseen task distributions (out-of-distribution). Through careful analyses, we show that for each of these two goals, current few-shot learning benchmarks have potential pitfalls in 1) performing model selection and hyperparameter tuning for a given meta-learning method and 2) comparing the performance of different meta-learning methods. Our results highlight that in order to reason about progress in this space, it is necessary to provide a clearer description of the goals of meta-learning, and to develop more appropriate corresponding evaluation strategies." @default.
- W3132000425 created "2021-03-01" @default.
- W3132000425 creator A5062151201 @default.
- W3132000425 creator A5072994032 @default.
- W3132000425 creator A5082543438 @default.
- W3132000425 date "2021-02-23" @default.
- W3132000425 modified "2023-09-27" @default.
- W3132000425 title "Lessons from Chasing Few-Shot Learning Benchmarks: Rethinking the Evaluation of Meta-Learning Methods." @default.
- W3132000425 hasPublicationYear "2021" @default.
- W3132000425 type Work @default.
- W3132000425 sameAs 3132000425 @default.
- W3132000425 citedByCount "0" @default.
- W3132000425 crossrefType "posted-content" @default.
- W3132000425 hasAuthorship W3132000425A5062151201 @default.
- W3132000425 hasAuthorship W3132000425A5072994032 @default.
- W3132000425 hasAuthorship W3132000425A5082543438 @default.
- W3132000425 hasConcept C111368507 @default.
- W3132000425 hasConcept C111919701 @default.
- W3132000425 hasConcept C119857082 @default.
- W3132000425 hasConcept C12725497 @default.
- W3132000425 hasConcept C127313418 @default.
- W3132000425 hasConcept C154945302 @default.
- W3132000425 hasConcept C162324750 @default.
- W3132000425 hasConcept C177264268 @default.
- W3132000425 hasConcept C187736073 @default.
- W3132000425 hasConcept C199360897 @default.
- W3132000425 hasConcept C2778572836 @default.
- W3132000425 hasConcept C2780451532 @default.
- W3132000425 hasConcept C2781002164 @default.
- W3132000425 hasConcept C41008148 @default.
- W3132000425 hasConcept C77967617 @default.
- W3132000425 hasConcept C8642999 @default.
- W3132000425 hasConceptScore W3132000425C111368507 @default.
- W3132000425 hasConceptScore W3132000425C111919701 @default.
- W3132000425 hasConceptScore W3132000425C119857082 @default.
- W3132000425 hasConceptScore W3132000425C12725497 @default.
- W3132000425 hasConceptScore W3132000425C127313418 @default.
- W3132000425 hasConceptScore W3132000425C154945302 @default.
- W3132000425 hasConceptScore W3132000425C162324750 @default.
- W3132000425 hasConceptScore W3132000425C177264268 @default.
- W3132000425 hasConceptScore W3132000425C187736073 @default.
- W3132000425 hasConceptScore W3132000425C199360897 @default.
- W3132000425 hasConceptScore W3132000425C2778572836 @default.
- W3132000425 hasConceptScore W3132000425C2780451532 @default.
- W3132000425 hasConceptScore W3132000425C2781002164 @default.
- W3132000425 hasConceptScore W3132000425C41008148 @default.
- W3132000425 hasConceptScore W3132000425C77967617 @default.
- W3132000425 hasConceptScore W3132000425C8642999 @default.
- W3132000425 hasLocation W31320004251 @default.
- W3132000425 hasOpenAccess W3132000425 @default.
- W3132000425 hasPrimaryLocation W31320004251 @default.
- W3132000425 hasRelatedWork W2117985513 @default.
- W3132000425 hasRelatedWork W2132836414 @default.
- W3132000425 hasRelatedWork W2401474588 @default.
- W3132000425 hasRelatedWork W2558073743 @default.
- W3132000425 hasRelatedWork W2558618579 @default.
- W3132000425 hasRelatedWork W2947012480 @default.
- W3132000425 hasRelatedWork W2952467940 @default.
- W3132000425 hasRelatedWork W2983100209 @default.
- W3132000425 hasRelatedWork W2996737117 @default.
- W3132000425 hasRelatedWork W2999930990 @default.
- W3132000425 hasRelatedWork W3034436181 @default.
- W3132000425 hasRelatedWork W3034692406 @default.
- W3132000425 hasRelatedWork W3035527463 @default.
- W3132000425 hasRelatedWork W3091492509 @default.
- W3132000425 hasRelatedWork W3091954049 @default.
- W3132000425 hasRelatedWork W3137020110 @default.
- W3132000425 hasRelatedWork W3173900519 @default.
- W3132000425 hasRelatedWork W3194253743 @default.
- W3132000425 hasRelatedWork W3198956694 @default.
- W3132000425 hasRelatedWork W944125612 @default.
- W3132000425 isParatext "false" @default.
- W3132000425 isRetracted "false" @default.
- W3132000425 magId "3132000425" @default.
- W3132000425 workType "article" @default.