Guide-yakabatana mibvunzo · Pakati Nhanho

Bias-Variance Tradeoff Quiz

Isa kurerekera, mutsauko, kugarisa, kubhegi, uye kupatsanurwa-kukanganisa uku uchiziva miganhu yekirasi yakaoma kunzwisisa curve.

Nzira dzekutungamira dzinoenderanaBias Variance Tradeoff
Mubvunzo 1 ye 8

Iyo modhi iri nyore kwazvo kubata iyo iri pasi peiti mu data inonyanya kutambura ne:

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Iyo modhi iri nyore kwazvo kubata iyo iri pasi peiti mu data inonyanya kutambura ne:
    • High musiyano
    • High bias
    • Low irreducible error
    • Kuwedzeredza
  2. Ndeipi mamiriro echiratidzo chekare chekusiyana kwepamusoro (kuwedzeredza)?
    • Kudzidzira uye bvunzo kukanganisa zvese zvakakwirira uye zvakafanana
    • Iyo modhi inofuratira akawanda maficha
    • Kuwedzera kurongeka kunowedzera kukanganisa kwekudzidziswa kune bvunzo kukanganisa
    • Kukanganisa kwekudzidziswa kwakadzikira zvakanyanya asi bvunzo kukanganisa kwakakwira zvakanyanya
  3. In the classic bias-variance picture described in the guide, which tradeoff can accompany increased model flexibility?
    • Kusarura kunoderera uye kusiyana kunowedzera
    • Zvose zvinowedzera
    • Ose ari maviri anoderera
    • Kusarura kunowedzera uye kusiyana kunoderera
  4. Kuwedzera L2 (ridge) kurongeka kune modhi kazhinji:
    • Inoderedza kusarura pamutengo wekusiyana kwakawanda
    • Inobvisa kukanganisa kusingagoneki
    • Inoderedza mutsauko pamutengo wekurerekera kwakawedzera zvishoma
    • Haina mhedzisiro pane bias kana musiyano
  5. Iko kusagadzirisika kukanganisa mukuora-kusiyana-siyana kunomiririra:
    • Kukanganisa kunokonzerwa nekusarudza iyo algorithm isiriyo
    • Kukanganisa kubva muruzha rwemukati mune data iyo isina modhi inogona kubvisa
    • Kukanganisa kunonyangarika nedata rakawanda rekudzidziswa
    • Kukanganisa kunokonzerwa chete nekusiyana kukuru
  6. Iyo 'double descent' chiitiko chinoonekwa mukudzidza kwakadzama chinoratidza kuti:
    • Kukanganisa kwekuyedza kunogona kudzikira zvakare sezvo mamodheru achikura kupfuura nzvimbo yekubatanidza data rekudzidziswa
    • Chikanganiso chekuyedza chinogara chichitevera iri nyore U-chimiro
    • Kusiyana hakuna basa kune neural network
    • Bias inogara iri zero mumhando dzakadzika
  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