Idanwo ti o sopọ mọ itọsọna · Alabọde Ipele

Idanwo Awọn Oṣuwọn Ẹkọ Cyclical

Checks triangular learning-rate cycles, the range test, momentum behavior and the difference between repeated cycles and one-cycle training.

Jẹmọ itọsọna awọn ọnaAwọn Iwọn Ẹkọ Yiyipo
Ibeere 1 ti 8

Iroye pataki wo ni awọn oṣuwọn ikẹkọ iyipo ni ipenija?

Awotẹlẹ gbogbo ibeere8 ibeere

Awọn idahun ati awọn alaye han bi o ṣe n mu ibeere naa.

  1. Iroye pataki wo ni awọn oṣuwọn ikẹkọ iyipo ni ipenija?
    • Wipe oṣuwọn ikẹkọ yẹ ki o dinku nigbagbogbo lakoko ikẹkọ
    • Iwọn ipele yẹn gbọdọ duro nigbagbogbo
    • Awọn awoṣe yẹn nilo awọn GPU
    • Ti data gbọdọ wa ni dapọ
  2. What did Smith report for cyclical learning rates on tested classification tasks?
    • They always lowered training loss to zero
    • They improved accuracy versus fixed rates, often in fewer iterations
    • They removed the need for validation data
    • They guaranteed escape from every sharp minimum
  3. Kini 'idanwo ibiti LR' ṣe?
    • Ṣe iwọn lilo iranti GPU
    • Kika awọn nọmba ti paramita awoṣe
    • Gba oṣuwọn ikẹkọ lọ si oke lakoko wiwo pipadanu lati mu awọn aala to dara
    • Ṣe idanwo awoṣe lori data pinpin kaakiri
  4. Ninu eto eto-ọmọ-ọkan, bawo ni iṣesi ṣe n huwa deede si iwọn ẹkọ?
    • Akoko duro titi jakejado
    • Ti ṣeto ipa-ọna si odo ni gbogbo akoko
    • Iṣaṣe pọsi ni deede ni igbesẹ pẹlu oṣuwọn naa
    • Ilọsiwaju n lọ ni idakeji - giga nigbati oṣuwọn ba lọ silẹ ati kekere nigbati oṣuwọn ba ga
  5. During a triangular cyclical learning-rate cycle, how does the rate move between its bounds?
    • It rises from the base rate to the maximum, then falls back toward the base
    • It only decreases from the first step
    • It stays fixed at the maximum
    • It jumps randomly at every minibatch
  6. Tani o ṣe afihan awọn oṣuwọn ikẹkọ iyipo ati idanwo ibiti LR?
    • Geoffrey Hinton
    • Leslie Smith
    • Yan LeCun
    • Andrej Karpathy
  7. What sets the duration of a cyclical learning-rate cycle?
    • The number of model layers
    • Awọn fokabulari iwọn
    • A chosen number of training iterations for the up-and-down pattern
    • The number of optimizer parameters
  8. What distinguishes the one-cycle policy from repeating several triangular cycles?
    • It keeps the learning rate constant throughout
    • It removes the optimizer’s momentum
    • It measures the learning rate only once
    • It uses one long learning-rate cycle for the training run