Guide-yakabatana mibvunzo · Pakati Nhanho

Adam uye Adaptive Optimizers Quiz

Inotarisa nguva dzaAdhamu gradient, kururamisa kwakarerekera, adaptive scaling, epsilon uye AdamW's decoupled huremu-kuora update. mune yakajairwa parameter update.

Nzira dzekutungamira dzinoenderanaAdam uye Adaptive Optimizers
Mubvunzo 1 ye 8

Ndeapi uwandu huviri hunotevedzwa naAdam paparameter yega yega?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Ndeapi uwandu huviri hunotevedzwa naAdam paparameter yega yega?
    • Avhareji inomhanya yemagradients uye yeskweya gradients
    • Iyo yakanyanya uye diki gradient yakamboonekwa
    • Chiyero chekudzidza uye saizi yebatch
    • Huremu kukosha uye chiratidzo chacho
  2. Nei Adhama achishandisa kururamisa kwakarerekera kune fungidziro yake yenguva?
    • Nekuti ma gradients anogara asina kunaka pakutanga mukudzidziswa
    • Nekuti maavhareji anofamba anotangira pa zero uye anorerekera ku zero pakutanga
    • Kuita kuti chiyero chekudzidza chiwedzere nekufamba kwenguva
    • Kubvisa outlier data mapoinzi kubva batch
  3. Ndeupi musiyano mukuru pakati paAdam naAdamW?
    • AdamW haishandisi kukurumidza zvachose
    • AdamW inobvisa fungidziro yenguva yechipiri
    • AdamW inobvisa uremu kuora kubva kune gradient-based update
    • AdamW inoshanda chete pane convolutional network
  4. Ibasa rei iro diki epsilon inobata mumutemo wekuvandudza waAdam?
    • Inoisa simba rekumhanya
    • Inodzora huwandu hwenguva dzekudzidziswa
    • Inowedzera gradient ukuru
    • Inodzivirira kupatsanurwa ne zero uye inodzikamisa iyo update
  5. Nei Adhamu achiwanzorondedzerwa se'anochinja'?
    • Iyo inopa zvinobudirira imwe neimwe parameter yayo yakayerwa nhanho saizi
    • Iyo inoshandura iyo network architecture panguva yekudzidziswa
    • Inogadzirisa iyo dataset kune modhi
    • Iyo inochinja pakati pechikamu uye regression otomatiki
  6. Which Adam parameter controls the exponential decay rate of its squared-gradient estimate?
    • Beta1
    • Beta2
    • Epsilon
    • The weight-decay coefficient
  7. A parameter repeatedly receives large gradients. Which Adam quantity helps scale down its effective step?
    • The number of model layers
    • The batch-normalization mean
    • The second-moment estimate of squared gradients
    • The AdamW weight-decay term alone
  8. What happens when beta2 is set closer to one, all else equal?
    • The optimizer stops using gradients
    • The first-moment estimate is deleted
    • Weight decay becomes decoupled automatically
    • The squared-gradient estimate decays more slowly and retains a longer history