Stochastic moyenne poids
Stochastic Weight Averaging (SWA) dafay jël moyenne bu yomb ci diisaayu model bi ci poñ yu bari ci diggu tàggat yaram, du tëye nataal bu mujj bi.
Résumé
This cheap trick often lands the model in a flatter, wider region of the loss landscape, which tends to generalize noticeably better on unseen data.
Plongeur bu xóot
Izmailov, Wilson ak ay naataango ñoo ko dugal ci 2018, SWA dafay jëfandikoo seetlu bi ñu gis ni SGD ak njàngum njàng mu sax wala njàngum siklik du jëm ci benn poñ - dafay rebondir ci wetu vale bu yaatu, bu dalal. Du tann benn ci barab yu bari yi, SWA dafay doxal njàngum njàng mu yéeg (muy faral di nekk wala cyclic) ci jamono yu mujj yi, ba noppi di moyenne poids yi mu dem, lu gëna bari ci jamono bu nekk. Poids moyenne yi ñoo gëna jege diggu gox bu dalal bi. Ndax lim yi ñuy jëfandikoo ngir normalisasioŋ ci lote yi dañu leen di xayma ci ay poñ yuñ tànn, SWA dafay laaj benn jéego bu gëna mag ci kaw done yi ngir xaymawaat BN biy daw ak variance yi ci model moyenne bi. Njëg li amul benn fayda, te njariñu njubte gi dafay méngoo ci tànneefi nataal yi ak feneen.
Gis-gis xarala
SWA dafay tëye ab moyenne buy daw w_SWA = (n·w_SWA + w_i)/(n+1) di yeesal sikl bu nekk, ci noonu la xeetu SGD biy dundu di wéy di jàngat ak tolluwaayu jàng bu yaatu. Moyenne ci espace poids mingi jege ensemble ci espace fonction waaye benn model la ci inference, du bari. Mekanism bi gëna am solo mooy minima yu plat yi dañu dëgër ci perturbation yu diis, moo tax surface yiy ñàkk tàggat/test ñu ngi nekk ci yoon, wàññi gap bi ci généralisation.
njeextalu pexe
Njëgg ak budget
Dogal yi architecture di jël dañuy indi njariñ ak njëgu liggéey bi ay at ci ginaaw.
dogal yu gëna leer
Njàngalem xarala yi dafay jàppale ekip yi ñu tànn li gën, te baña yam ci li gëna bees daal.
Xool kalite
Tanneef yu gëna baax ci wàllu ingeñër dina wàññi jafe-jafe yi ci wàllu wóor ci liggéey bi.
Ëlëgu moyenne poids stochastic
SWA jural na ay anam yu melni SWA-Gaussian (SWAG) ngir ñàkka wóor gu Bayesian, te xalaatu moyenne bi leegi dafay jàppale moyenne exponentielle bi ñuy jëfandikoo bu bari ci xeetu diffusion, jàng bu ñuy saytu seen bopp, ak model bu mag bu ñuy njëkka tàggat. Xaarandil ni diisaay bi dafay des ci 'lunch bu amul fey' ci rëset yi ñuy tàggat, ak gëstu yu koy yokk ngir boole xeetu tàggat yu moom seen bopp (suppu model) ak gëna suqali etalonnage ci wetu njubte bu ñor.
Doxal ci àdduna dëgg
Yokkateg njubteg test bu ResNet ak DenseNet ci CIFAR ak ImageNet te doo fay dara.
SWAG (SWA-Gaussian) defar xayma yuñ kalibree ci ñàkka wóor ngir xam luy waaja am ci benn tàggat yaram.
EMA-ci-poids yiy dakkal reso échantillonnage ci defarkati nataali diffusion yu melni Diffusion bu dëgër.
Tabax 'model supps' ci moyenne checkpoint yu bari yuñ defar bu baax ngir gëna am doole te doo tàggataat.
Risk yi ak balustrade yi
Optimize benn benchmark mën na nëbb ñakk kattan yu gëna yaatu ci sistem bi.
Njëg li ñuy fay ci infrastructure yi ak ci toppatoo dañuy faral di suufeel.
Bu sistem yi di gëna xawa jafee xam, jafe-jafe yi am ci wàllu kaaraange ak seetlu mën nañu gëna bari.
Roadmap ngir samp gi
Mandargal latency, kalite, ak njëg yi laata ngay jëfandikoo.
Benchmark ci biir sargal ak done yu dëggu.
Jumtukaay bi di saytu njuumte yi, derive bi ak njeextalu jëfandikukat bi.
Waajal rollback ak yooni tontu ci jafe-jafe yi laata ngay eskale.
Weyal di banneexu
Free newsletter
Get the daily AI briefing
Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.
One email each weekday. Unsubscribe in one click. We never sell or share your address.
Test yourself
Take the Stochastic Weight Averaging quiz
Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.
Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation
Gis bi ci topp
Tambali poid
Laaj yi ñuy faral di laaj
What is Stochastic Weight Averaging?
Stochastic Weight Averaging (SWA) dafay jël moyenne bu yomb ci diisaayu model bi ci poñ yu bari ci diggu tàggat yaram, du tëye nataal bu mujj bi. Kaf gu yomb gii dafay faral di teg model bi ci gox bu gëna dalal, bu gëna yaatu ci paysage perte bi, te loolu dafay gëna generalise ci done yuñu gisul.
Lan mooy moyenne poids stochastic ci dëgg?
SWA dafay moyenne parametre model yi (poids) yiñ dajale ci yenn checkpoints ci diiru fase bu mujj ci tàggat bi, du prediction wala gradient.
Lan moo waral SWA di gëna suqali généralisasioŋ bi?
Moyenne dafay toxal solution bi ci diggu gox bu dalal ci surface perte bi, ba noppi minima yu plat yi dañuy gëna generalise ndax train bi ak perte test yi dañuy des ci yoon wi.
Ban jéego bu gëna am solo la SWA soxla ci reso yiy jëfandikoo normalisasioŋ ci lots?
Ndax lim BN a ngi aju ci poid yi, model moyenne bi soxla benn pass bu gëna bari ci done yi ngir xaymawaat moyenne yi ak variance yi.
Ban jeffin juy jàng mooy gëna bari luñu koy jëfandikoo ci wàllu moyenne SWA?
SWA dafay tëye tolluwaayu jàng bu yéeg bu baax wala buy wër, moo tax SGD dafay wéy di jàngat gox bu yaatu bi ñuy jël poñ yi ci moyenne.
Lan mooy njëgu inference bu SWA buñu ko méngale ak ensemble bu cosaan bu N models?
SWA dafay dindi ay checkpoint yu bari ci benn ensemble poid moyen, kon inference bi dafay njëg benn model te du N.