Ikibazo gihuza ibibazo · Hagati Urwego

Ikibazo Cyamanuka Ikibazo

Ongera usuzume icyerekezo gisanzwe-cyerekezo, icyerekezo-cyo kwiga intambwe, icyitegererezo cya gradients, umuvuduko, Guhuza kwa Adamu, kuvugurura imibare, hamwe nimbibi zihagarara.

Inzira zijyanye no kuyoboraKumanuka
Ikibazo 1 Bya 8

In plain gradient descent, which direction is used relative to the objective's gradient?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. In plain gradient descent, which direction is used relative to the objective's gradient?
    • The negative-gradient direction
    • The positive-gradient direction
    • An arbitrary perpendicular direction by definition
    • A direction chosen without evaluating any gradient
  2. Igipimo cyo kwiga kigenzura iki?
    • Umubare wibice muri moderi
    • Nibyiciro bingahe icyitegererezo gishobora guhanura
    • Ingano ya buri ntambwe yatewe mugihe cyo gukora neza
    • Umubare w'amahugurwa yakoreshejwe
  3. How does a mini-batch gradient relate to the full selected training-data objective?
    • It uses no examples
    • It is computed on the batch and used as an estimate of the full-data gradient
    • It must equal the full-data gradient on every step
    • It only exists for images
  4. Ni ikihe kibazo igipimo cyo kwiga ari kinini cyane?
    • Imyitozo iba buhoro cyane ariko ihamye
    • Icyitegererezo cyirengagije icyiciro
    • Dataset ihinduka ruswa
    • Icyitegererezo kirenga byibuze kandi gishobora gutandukana
  5. What information does momentum carry into later gradient-descent updates?
    • Only the model's vocabulary size
    • A permanent deletion of all earlier gradients
    • A running weighted combination of earlier gradients
    • An unrelated random reset of every weight
  6. Ni ibihe bintu by'ingenzi biranga Adam optimizer ugereranije no kumanuka gahoro gahoro?
    • Ihuza igipimo cyiza cyo kwiga kuri buri kintu
    • Ikoresha intambwe imwe ihamye kuri buri kintu cyose
    • Ntabwo bisaba gradients
    • Itoza gusa ibisohoka
  7. For plain gradient descent with old weight two, gradient three, and learning rate one-tenth, what is the new weight?
    • Two-point-three
    • One-point-nine-seven
    • Bitatu
    • One-point-seven
  8. What does a zero gradient fail to prove for a nonconvex training objective?
    • That the gradient value is zero at the evaluated point
    • That the current point is a global minimum
    • That the objective has been evaluated locally
    • That a stationary-point condition holds