Maswali yanayounganishwa na mwongozo · Kati Kiwango

Lebo ya Maswali ya Kulaini

Inashughulikia shabaha zilizolainishwa na epsilon, athari za urekebishaji zinazotegemea kazi, mabadiliko ya kunereka, na kwa nini matokeo hutegemea data na lengo.

Njia za mwongozo zinazohusianaLebo ya Smoothing
Swali 1 ya 8

Je, urekebishaji wa lebo hubadilika nini kuhusu malengo ya mafunzo?

Hakiki kila swali8 maswali

Majibu na maelezo huonekana unapojibu maswali.

  1. Je, urekebishaji wa lebo hubadilika nini kuhusu malengo ya mafunzo?
    • Inapunguza malengo ya moto-moja ili darasa la kweli liwe chini ya 1.0 na zingine ziko juu kidogo ya 0
    • Inaondoa kazi ya kupoteza
    • Inafuta madarasa yasiyo sahihi
    • Inaongeza kiwango cha kujifunza mara mbili
  2. Thamani ya kawaida ya parameta ya kulainisha epsilon ni:
    • 0.9
    • 0.1
    • 5.0
    • 0.0
  3. What is one intended effect of replacing hard one-hot targets with smoothed targets?
    • Force every class to have equal probability
    • Guarantee the largest possible logit for the true class
    • Avoid requiring the labeled class to receive a target probability of exactly one
    • Remove the need for labeled training data
  4. Which landmark architectures are prominent examples of label-smoothing use?
    • AlexNet and LeNet only
    • ResNet-50 and VGG only
    • AlphaGo na AlphaZero
    • Kuanzishwa-v3 na Transformer
  5. What can label smoothing do to calibration?
    • It can improve calibration in some settings, but the effect is task- and model-dependent
    • It guarantees perfect calibration for every classifier
    • It cannot affect predicted probabilities
    • It always makes confidence equal accuracy
  6. Je, ni hasara gani iliyorekodiwa ya kulainisha lebo?
    • Daima hupunguza usahihi
    • Inaweza kufuta ulinganifu mzuri kati ya darasa, kuumiza kunereka kwa maarifa
    • Inahitaji data ya mafunzo mara mbili
    • Inafanya kazi kwa maandishi tu
  7. Why can label smoothing complicate knowledge distillation?
    • It converts every class label into an image
    • It forces the student to use a different tokenizer
    • It can remove information in logits about similarities among incorrect classes
    • It prevents the teacher from producing logits
  8. What should a practitioner verify when comparing label-smoothing implementations?
    • Whether the model has a web browser
    • Whether training uses a validation filename
    • Whether logits are stored as text
    • Whether epsilon is distributed over all classes or only the non-target classes