Maswali yanayounganishwa na mwongozo · Kati Kiwango

Maswali ya Kushuka kwa Gradient

Kagua mwelekeo wa mteremko wa kushuka, hatua za kiwango cha ujifunzaji, viwango vya upinde wa mvua vilivyotolewa sampuli, kasi, urekebishaji wa Adamu, masasisho ya nambari, na vikomo vya pointi zisizobadilika.

Njia za mwongozo zinazohusianaKushuka kwa Gradient
Swali 1 ya 8

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

Hakiki kila swali8 maswali

Majibu na maelezo huonekana unapojibu maswali.

  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. Je, kiwango cha kujifunza kinadhibiti nini?
    • Idadi ya tabaka katika mfano
    • Mfano unaweza kutabiri madarasa ngapi
    • Ukubwa wa kila hatua iliyochukuliwa wakati wa uboreshaji
    • Kiasi cha data ya mafunzo iliyotumika
  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. Je, kiwango cha kujifunza ambacho ni kikubwa sana kinaweza kusababisha tatizo gani?
    • Mafunzo yanakuwa polepole sana lakini thabiti
    • Mfano hupuuza upinde rangi
    • Seti ya data inaharibika
    • Mfano huo unazidi kiwango cha chini na unaweza kutofautiana
  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 kipengele gani muhimu cha kiboreshaji cha Adamu ikilinganishwa na asili ya upinde rangi wazi?
    • Inabadilisha kiwango cha ujifunzaji bora kwa kila kigezo
    • Inatumia hatua moja maalum kwa kila parameta
    • Haihitaji gradients
    • Inafunza safu ya pato tu
  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
    • Tatu
    • 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