Rehber bağlantılı sınav · Orta Seviye

Geriye Yayılım Testi

Geri yayılma gradyanlarını, ters bağımlılık sırasını, önbelleğe alınmış ileri değerleri, zincir kuralı çarpımını, paylaşılan yol toplamlarını ve optimize edici ayrımını test edin.

İlgili kılavuz yollarıGeri yayılım
Soru 1 arasında 8

What quantity does backpropagation compute for a neural network's relevant weights?

Her soruyu önizleyin8 sorular

Testi çözdüğünüzde cevaplar ve açıklamalar görünür.

  1. What quantity does backpropagation compute for a neural network's relevant weights?
    • Derivatives of the loss with respect to those weights
    • A causal percentage of total error uniquely assigned to each weight
    • The architecture drawing without any calculation
    • A permanently updated set of weights without an optimizer
  2. Geri yayılımın merkezinde hangi matematik kuralı yer alır?
    • Pisagor teoremi
    • Bayes teoremi
    • Analizin zincir kuralı
    • İkinci dereceden formül
  3. Why can reverse-mode backpropagation compute many parameter derivatives efficiently?
    • It avoids all local derivatives
    • It reuses graph structure and relevant intermediate results
    • It calculates only the output prediction again
    • It guarantees every implementation has exactly one forward pass of runtime
  4. Geri yayılım gradyanları hangi sırayla hesaplar?
    • Giriş katmanından çıkış katmanına doğru
    • Tüm katmanlar, sıra olmadan aynı anda
    • Yalnızca ortadaki gizli katman için
    • Çıkış katmanından geriye doğru giriş katmanına doğru
  5. Does computing backpropagation gradients by itself establish that model weights were updated?
    • Yes; every backward computation automatically applies a learning rate
    • Yes; gradient values always replace the weights directly
    • No; the optimizer update is a separate step
    • Only if the network has no parameters
  6. In a conventional cached-backpropagation workflow, what does the forward computation provide for later derivative calculations?
    • Relevant intermediate values that can be retained for the backward computation
    • A completed optimizer update without any update rule
    • A guarantee that all later evaluation errors disappear
    • A permanent deletion of the training objective
  7. On one composed backpropagation path, local scalar derivatives are three and four. What is their chain-rule product?
    • Yedi
    • Bir
    • Three quarters
    • On iki
  8. A shared parameter has loss-gradient contributions two and five through two paths. What combined contribution is accumulated?
    • On
    • Yedi
    • üç
    • Two fifths