Gwara reVisual AI

Wasserstein GAN

Wasserstein GAN (WGAN) igadziriso yechinangwa cheGAN chekudzidzisa chinoshandisa chinhambwe cheWasserstein pachinzvimbo chekutanga min-max kurasikirwa.

Teerera kugwaro iriAI-yakarondedzerwa · ~2 min

Tanga kuverengaTora mubvunzo
Pa peji ino2 min verenga
  1. Pfupiso
  2. Kudzika Kwakadzika
  3. Kuitwa muNyika yeChokwadi
  4. Ramangwana reWasserstein GAN
  5. Kuzvishandisa
  6. Ramba Uchiongorora
  7. Mibvunzo inowanzo bvunzwa

Pfupiso

Inoita kuti kudzidziswa kweGAN kuzivikanwe zvakanyanya kuvimbika uye kunopa kukosha kwekurasikirwa kunowirirana nemhando yemufananidzo.

Kudzika Kwakadzika

MaGAN ekutanga anodzidzisa manetwork maviri mukudhonzana: jenareta rinogadzira mifananidzo yemanyepo uye munhu anosarura anoedza kuzviona. Izvi zvinowanzopunzika kana kumira nekuti kurasikirwa kwemusarura hakurevi chinhu chinobatsira nezvebudiriro. WGAN, yakaunzwa naArjovsky, Chintala, naBottou muna 2017, inotsiva musarura ne 'mutsoropodzi' anotarisisa kuti mufananidzo unotaridzika sei pachiyero chinoenderera pane kurongedza real-vs-fake. Chinangwa chekudzidziswa chinova Wasserstein (pasi-mover's) chinhambwe pakati peiyo chaiyo uye inogadzirwa data kugovera. Chinhambwe ichi chinopa akapfava, ane chirevo magradients kunyangwe iwo maviri magoveta asingadhure, zvinoshamisa kuderedza modhi kudonha uye kuita iyo yekurasika curve chiratidzo chechokwadi chemhando.

Kuitwa muNyika yeChokwadi

Kugadzira zviso zvephotorealistic uye maumbirwo apo vanilla GANs akadonha kune mashoma akadzokororwa kubuda

Kugadzira mifananidzo yekurapa yekugadzira, yakadai seMRI kana histology zvigamba, kuwedzera mashoma akanyorwa dataset.

Kuenzanisira particle-kudhumhana zviitiko muhigh-energy fizikisi simulations uko kudzidziswa kwakagadzikana kwakakosha

Kushanda seyekutanga bhenji mukutsvaga kweML nekuti kurasikirwa kwayo kunoteedzera mhando yemhando pamusoro pekudzidziswa

Ramangwana reWasserstein GAN

WGAN's musimboti nzwisiso, iyo sarudzo yekugovera chinhambwe inoumba gradient mhando, ichiri inonzwika kuburikidza nekugadzira modhi. Nepo mamodheru ekuparadzira ave kutonga kuumbwa kwemifananidzo, mazano akanyanya-yekutakura kubva kuWGAN anoonekwazve mukuyerera, nzira dzeSchrodinger-bhiriji, uye distillation yemamodhi ekuparadzira kuita mashoma-nhanho jenareta. Tarisira zvinangwa zveWasserstein-style zvekuramba uchizivisa nzira dzakasanganiswa uko kudzidziswa kwakagadzikana uye kurasikirwa kunokosha metric nyaya, kunyanya mune zvesainzi uye yakaderera-data.

Kuzvishandisa

Nei zvichikosha

  • Kumhanya uye chiyero. Visual AI inogona kuita otomatiki yekuongorora, yekuona, uye yekumaka mabasa pachiyero.
  • Vaka sarudzo. Zvikwata zvekugadzira zvinogona prototype pfungwa nekukurumidza nekudzokororwa kwemaoko mashoma.
  • Team uye workflow. Mashandisirwo anogona kushandisa masaini emifananidzo nemavhidhiyo ayo aimbove akaoma kugadzirisa.

Chenjera

  • Kodzero dzemifananidzo uye kubvumirwa kunogona kuve njodzi dzepamutemo kana provenance isina kujeka.
  • Kuita kwemuenzaniso kunogona kusiyanisa kupenya, huwandu hwevanhu, uye nharaunda.
  • Manyepo enhema anogona kusacherechedzwa kunze kwekunge zvikumbaridzo zvekuvimba zvikatariswa.

Matanho anotevera

  1. Tsanangura maitiro ekugamuchirwa echokwadi, kurangarira, uye mutengo wekukanganisa.
  2. Edzai nedata rinoenderana nemamiriro chaiwo ekugadzira.
  3. Wedzera ongororo yemunhu kune yakaderera-kusavimbika kana yakakwirira-inokanganisa kufanotaura.
  4. Tevera modhi kudonha uye simbisa mushure mekuchinja kwekamera kana dataset.

Ramba Uchiongorora

Kuona kweKomputa

Nzwisisa masisitimu epasi anopa simba kuona AI.Verenga Guide

AI Kugadzira mifananidzo

Ongorora kugadzira mafambiro uye modhi tradeoffs.Verenga Guide

Tsamba yemahara

Tora iyo yemazuva ese AI pfupiso

Nhatu dzakasimbiswa dzeAI nyaya mangwanani ega ega pakati pevhiki, dzakanyorwa neChirungu chakajeka. Yemahara zvachose, hapana ads.

Imwe email zuva rega rega revhiki. Bvisa kunyoresa kamwe chete. Hatimbotengese kana kugovera kero yako.

Zviedze iwe pachako

Take the Wasserstein GAN quiz

Mhinduro yepakarepo pamhinduro yega yega, uye chitupa chinogovaniswa chine ID inobatika kana wapedza kosi.

Tanga mibvunzo

Tsigira dzidzo yemahara yeAI. AI Understanding ndeye 501(c)(3) isingabatsiri - hapana ads, hapana paywall, zvachose. Ita chipo

Mibvunzo inowanzo bvunzwa

Chii chinonzi Wasserstein GAN?

Wasserstein GAN (WGAN) igadziriso yechinangwa cheGAN chekudzidzisa chinoshandisa chinhambwe cheWasserstein pachinzvimbo chekutanga min-max kurasikirwa. Inoita kuti kudzidziswa kweGAN kuzivikanwe zvakanyanya kuvimbika uye kunopa kukosha kwekurasikirwa kunowirirana nemhando yemufananidzo.

Ndeapi mametric anoshandiswa neWGAN kuenzanisa kugovera chaiko uye kwakagadzirwa?

WGAN inotsiva iyo yekutanga Jensen-Shannon-yakavakirwa chinangwa neiyo Wasserstein chinhambwe, chinopa akapfava magradients kunyangwe kugovera kusingadhure.

MuWGAN, network yaive isarura inotumidzwa zita rekuti chii, uye nei?

Mutsoropodzi anokora mifananidzo pachiyero chinoenderera pane kurongedza chaiyo vs fake, izvo zvinoita kuti chinangwa cheWasserstein chishande.

Sei mutsoropodzi weWGAN achifanira kumanikidzwa kuve 1-Lipschitz?

Hunyambiri hunoita kuti WGAN ifungidzire kureba kweWasserstein inongobata mabasa e1-Lipschitz, saka magidhi emutsoropodzi anofanira kusungwa.

Iyo yekutanga WGAN yakasimbisa sei iyo Lipschitz kumanikidza?

Pepa rekutanga reWGAN rakashandisa crude weight clipping; WGAN-GP yakazoitsiva nechirango chakapfava.

Ndeipi dambudziko iro WGAN inoderedza zvakanyanya kana ichienzaniswa nevanilla GAN?

WGAN's yakapfava gradients curb mode inodonha uye kuburitsa kurasikirwa kunonyatso kuenderana nemhando yemuenzaniso.

Ramba uchidzidza

Mazano anoenderana

Mamwe madhairekitori akasarudzirwa nyaya iyi

Wana chitupa cheAI chemaharaKosi pfupi yemahara nebvunzo. Hapana mutengo, hapana kadhi. Inoda kusvika maawa 2.5.Tanga mahara →