Gids-gekoppelde quiz · Middelmatig Niveau
Backpropagatie Quiz
Test backpropagation-gradiënten, omgekeerde afhankelijkheidsvolgorde, in de cache opgeslagen voorwaartse waarden, vermenigvuldiging van ketenregels, sommen op gedeelde paden en scheiding van de optimalisatie.
Vraag 1 van 8
What quantity does backpropagation compute for a neural network's relevant weights?
Bekijk elke vraag8 vragen
Antwoorden en uitleg verschijnen terwijl je de quiz doet.
- 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
- Welke wiskundige regel vormt de kern van backpropagation?
- De stelling van Pythagoras
- De stelling van Bayes
- De kettingregel van calculus
- De kwadratische formule
- 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
- In welke volgorde berekent backpropagation gradiënten?
- Van de invoerlaag naar de uitvoerlaag
- Alle lagen tegelijkertijd zonder volgorde
- Alleen voor de middelste verborgen laag
- Van de uitvoerlaag achteruit naar de invoerlaag
- 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?
- Zeven
- Eén
- Three quarters
- Twelve
- A shared parameter has loss-gradient contributions two and five through two paths. What combined contribution is accumulated?
- Tien
- Zeven
- Drie
- Two fifths