Ikibazo gihuza ibibazo · Hagati Urwego

Kubogama-Gutandukanya Ubucuruzi

Koresha kubogama, gutandukana, guhora, gutekera, hamwe na kwaduka-kwibeshya kwangirika mugihe umenye imipaka ya classique igoye.

Inzira zijyanye no kuyoboraKubogama Kubi
Ikibazo 1 Bya 8

Icyitegererezo cyoroshye cyane gufata uburyo bwibanze mumibare yababajwe cyane:

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Icyitegererezo cyoroshye cyane gufata uburyo bwibanze mumibare yababajwe cyane:
    • Itandukaniro ryinshi
    • Kubogama cyane
    • Ikosa rito ridasubirwaho
    • Kurenza urugero
  2. Ni ibihe bintu ni ikimenyetso cyerekana itandukaniro ryinshi (birenze)?
    • Amahugurwa hamwe nikizamini cyo hejuru byombi kandi birasa
    • Icyitegererezo cyirengagije ibintu byinshi
    • Ongeraho byinshi mubisanzwe byongera ikosa ryamahugurwa kugana ikizamini
    • Ikosa ryamahugurwa ni rito cyane ariko ikosa ryikizamini riri hejuru cyane
  3. In the classic bias-variance picture described in the guide, which tradeoff can accompany increased model flexibility?
    • Kubogama kugabanuka no gutandukana biriyongera
    • Byombi biriyongera
    • Byombi biragabanuka
    • Kubogama biriyongera kandi itandukaniro rigabanuka
  4. Ongeraho L2 (ridge) isanzwe kurugero muri rusange:
    • Kugabanya kubogama kubiciro byinyuranye
    • Kurandura ikosa ridasubirwaho
    • Kugabanya itandukaniro kubiciro byo kubogama gato
    • Nta ngaruka bigira kubogama cyangwa gutandukana
  5. Ikosa ridasubirwaho muri bias-variance decomposition ryerekana:
    • Ikosa ryatewe no guhitamo algorithm itari yo
    • Ikosa riva mu rusaku rwihariye mu makuru nta moderi ishobora gukuraho
    • Ikosa ryabuze hamwe namakuru menshi yo guhugura
    • Ikosa ryatewe gusa no gutandukana cyane
  6. Ibintu 'bibiri kumanuka' byagaragaye mu myigire yimbitse byerekana ko:
    • Ikosa ryikizamini rirashobora kugabanuka uko moderi ikura kure cyane yingingo yo guhuza amakuru yamahugurwa
    • Ikosa ryikizamini buri gihe rikurikiza U-shusho yoroshye
    • Ibinyuranyo ntabwo bigira akamaro kumurongo wimitsi
    • Kubogama buri gihe ni zeru muburyo bwimbitse
  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