quiz bu lëkkale ak tegtal · Diggante Niveau

Laaj ak njaay ci wàllu wuute

Jëfandikool njaaxaanaay, variance, yamale, sac, ak decomposition erreur carré booba ngay xàmmee yam ci courbe complexité bu yàgg bi.

Yooni njiit yu jëm cikompromis variaasioŋ bias
Laaj 1 de 8

Benn model bu yomb lool ngir jàpp motif bi ci suufu done yi dafay njëkka am jafe-jafe:

Xoolaatal laaj bu nekk8 laaj

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

  1. Benn model bu yomb lool ngir jàpp motif bi ci suufu done yi dafay njëkka am jafe-jafe:
    • Variance bu rëy
    • Bias bu rëy
    • Jumtukaay bu ndaw buñu mënatul wàññi
    • Soppi
  2. Ban senaario mooy màndarga biñ gëna xam ci variance bu rëy (jëfandikoo lu ëpp)?
    • Jumtukaay ci tàggat yaram ak ci test yépp dañu yéeg te dañuy nuru
    • Modèle bi dafay baña bayyi xel ci man-man yi gëna bari
    • Yokk yeneen yamale dafay yokk njuumti tàggat yaram ci njuumti test
    • Jumtukaayu tàggat yaram dafa tuuti lool waaye njuumti ci test bi dafa gëna rëy
  3. In the classic bias-variance picture described in the guide, which tradeoff can accompany increased model flexibility?
    • Bias dafay wàññeeku, variance di yokk
    • ñoom ñaar ñu ngi yokk
    • ñoom ñaar ñu ngi wàññeeku
    • Bias yokk te variance wàññeeku
  4. Yokk L2 (ridge) yamale ci benn model ci anam wu yaatu:
    • Wàññi njaaxaanaay ci njëgu faraas bu gëna bari
    • Dafay dindi njuumte buñu mënul wàññi
    • Wàññi variance ci njëgu tuuti bias
    • Amul benn njeexital ci njuumte wala variance
  5. Jumtukaay bi mënul wàññeeku ci xaaj bias-variance dafay màndargaal:
    • Jumtukaay bi juddoo ci tànneef algorithm bu jaarul yoon
    • Jumtukaay bu bawoo ci bruit bi nekk ci done yi te benn model mënu ko dindi
    • Jumtukaay buy reer ak yeneen done yuñ tàggat
    • Jumtukaay bi variance bu rëy rekk moo ko waral
  6. Fenomen 'ñaari wàcci' yi ñuy seetlu ci jàngat bu xóot bi dafay wane ni:
    • Jumtukaayu test mën na wàññeekuwaat ginaaw bi model yi màggee fu sori ba weesu poñ bi ñuy boole done yiñ tàggat
    • Jumtukaayu test dafay topp formu U bu yomb
    • Variance amul benn njariñ ci reso neuronal yi
    • Bias du saa yu nekk nul ci model yu xóot yi
  7. In a squared-error decomposition, squared bias is 2, variance is 3, and irreducible noise is 1. What is the total expected error?
    • 14
    • 5
    • 3
    • 6
  8. Why can averaging many de-correlated trees help in the random-forest example?
    • It removes all noise from the data-generating process
    • It guarantees zero test error
    • It can reduce the variance of the combined prediction
    • It converts every tree into a linear model