Matches in SemOpenAlex for { <https://semopenalex.org/work/W4365597197> ?p ?o ?g. }
- W4365597197 abstract "Machine learning (ML) models can fail in unexpected ways in the real world, but not all model failures are equal. With finite time and resources, ML practitioners are forced to prioritize their model debugging and improvement efforts. Through interviews with 13 ML practitioners at Apple, we found that practitioners construct small targeted test sets to estimate an error's nature, scope, and impact on users. We built on this insight in a case study with machine translation models, and developed Angler, an interactive visual analytics tool to help practitioners prioritize model improvements. In a user study with 7 machine translation experts, we used Angler to understand prioritization practices when the input space is infinite, and obtaining reliable signals of model quality is expensive. Our study revealed that participants could form more interesting and user-focused hypotheses for prioritization by analyzing quantitative summary statistics and qualitatively assessing data by reading sentences." @default.
- W4365597197 created "2023-04-15" @default.
- W4365597197 creator A5006087928 @default.
- W4365597197 creator A5019311613 @default.
- W4365597197 creator A5034842879 @default.
- W4365597197 creator A5068961297 @default.
- W4365597197 creator A5073527485 @default.
- W4365597197 date "2023-04-19" @default.
- W4365597197 modified "2023-10-18" @default.
- W4365597197 title "Angler: Helping Machine Translation Practitioners Prioritize Model Improvements" @default.
- W4365597197 cites W1552182777 @default.
- W4365597197 cites W2098865355 @default.
- W4365597197 cites W2108082276 @default.
- W4365597197 cites W2108816886 @default.
- W4365597197 cites W2129905273 @default.
- W4365597197 cites W2147192413 @default.
- W4365597197 cites W2165558283 @default.
- W4365597197 cites W2250342921 @default.
- W4365597197 cites W2282821441 @default.
- W4365597197 cites W2294798173 @default.
- W4365597197 cites W2330219538 @default.
- W4365597197 cites W2751722336 @default.
- W4365597197 cites W2756888147 @default.
- W4365597197 cites W2757188127 @default.
- W4365597197 cites W2760327630 @default.
- W4365597197 cites W2807434717 @default.
- W4365597197 cites W2898962441 @default.
- W4365597197 cites W2902180983 @default.
- W4365597197 cites W2918341242 @default.
- W4365597197 cites W2922349260 @default.
- W4365597197 cites W2948897128 @default.
- W4365597197 cites W2949858875 @default.
- W4365597197 cites W2952328691 @default.
- W4365597197 cites W2963198001 @default.
- W4365597197 cites W2963214037 @default.
- W4365597197 cites W2963323070 @default.
- W4365597197 cites W2963457723 @default.
- W4365597197 cites W2963526187 @default.
- W4365597197 cites W2963532001 @default.
- W4365597197 cites W2964130895 @default.
- W4365597197 cites W2970442950 @default.
- W4365597197 cites W2970641574 @default.
- W4365597197 cites W2971141904 @default.
- W4365597197 cites W2980941473 @default.
- W4365597197 cites W2985337504 @default.
- W4365597197 cites W3004542466 @default.
- W4365597197 cites W3012624518 @default.
- W4365597197 cites W3031531573 @default.
- W4365597197 cites W3035408261 @default.
- W4365597197 cites W3035507081 @default.
- W4365597197 cites W3038035611 @default.
- W4365597197 cites W3092425241 @default.
- W4365597197 cites W3103029570 @default.
- W4365597197 cites W3105928338 @default.
- W4365597197 cites W3125603643 @default.
- W4365597197 cites W3135453287 @default.
- W4365597197 cites W3159892921 @default.
- W4365597197 cites W3171654528 @default.
- W4365597197 cites W3174057701 @default.
- W4365597197 cites W3175487198 @default.
- W4365597197 cites W3184545100 @default.
- W4365597197 cites W3184924454 @default.
- W4365597197 cites W3204393347 @default.
- W4365597197 cites W3205290952 @default.
- W4365597197 cites W3213728774 @default.
- W4365597197 cites W4200296112 @default.
- W4365597197 cites W4214895166 @default.
- W4365597197 cites W4223555496 @default.
- W4365597197 cites W4224992683 @default.
- W4365597197 cites W4234651772 @default.
- W4365597197 cites W4243863038 @default.
- W4365597197 cites W4245267204 @default.
- W4365597197 cites W4283269716 @default.
- W4365597197 cites W4285113702 @default.
- W4365597197 cites W4288359825 @default.
- W4365597197 cites W4298326701 @default.
- W4365597197 cites W4313068521 @default.
- W4365597197 doi "https://doi.org/10.1145/3544548.3580790" @default.
- W4365597197 hasPublicationYear "2023" @default.
- W4365597197 type Work @default.
- W4365597197 citedByCount "0" @default.
- W4365597197 crossrefType "proceedings-article" @default.
- W4365597197 hasAuthorship W4365597197A5006087928 @default.
- W4365597197 hasAuthorship W4365597197A5019311613 @default.
- W4365597197 hasAuthorship W4365597197A5034842879 @default.
- W4365597197 hasAuthorship W4365597197A5068961297 @default.
- W4365597197 hasAuthorship W4365597197A5073527485 @default.
- W4365597197 hasBestOaLocation W43655971971 @default.
- W4365597197 hasConcept C107457646 @default.
- W4365597197 hasConcept C111472728 @default.
- W4365597197 hasConcept C111919701 @default.
- W4365597197 hasConcept C119857082 @default.
- W4365597197 hasConcept C127413603 @default.
- W4365597197 hasConcept C138885662 @default.
- W4365597197 hasConcept C154945302 @default.
- W4365597197 hasConcept C168065819 @default.
- W4365597197 hasConcept C17744445 @default.
- W4365597197 hasConcept C195094911 @default.
- W4365597197 hasConcept C199360897 @default.
- W4365597197 hasConcept C199539241 @default.