Quiz lié à un guide · Moyen Niveau
Quiz sur la rétropropagation
Testez les gradients de rétropropagation, l'ordre de dépendance inverse, les valeurs avancées mises en cache, la multiplication des règles de chaîne, les sommes de chemin partagé et la séparation de l'optimiseur.
Question 1 de 8
What quantity does backpropagation compute for a neural network's relevant weights?
Prévisualisez chaque question8 des questions
Les réponses et les explications apparaissent au fur et à mesure que vous répondez au quiz.
- 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
- Quelle règle mathématique est au cœur de la rétropropagation ?
- Le théorème de Pythagore
- Théorème de Bayes
- La règle en chaîne du calcul
- La formule quadratique
- 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
- Dans quel ordre la rétropropagation calcule-t-elle les gradients ?
- De la couche d'entrée vers la couche de sortie
- Toutes les couches simultanément sans ordre
- Uniquement pour la couche cachée du milieu
- De la couche de sortie vers l'arrière vers la couche d'entrée
- 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
- 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
- On one composed backpropagation path, local scalar derivatives are three and four. What is their chain-rule product?
- Sept
- Un
- Three quarters
- Twelve
- A shared parameter has loss-gradient contributions two and five through two paths. What combined contribution is accumulated?
- Dix
- Sept
- Trois
- Two fifths