Ikibazo gihuza ibibazo · Hagati Urwego

Ikibazo cya Autoencoders

Kugenzura intego zo kwiyubaka, kwikuramo rwihishwa, kwerekana, ibiranga umurongo no gutandukanya ibinyabiziga bisanzwe kandi bitandukanye.

Inzira zijyanye no kuyoboraAutoencoders
Ikibazo 1 Bya 8

Niyihe ntego ya 'icupa' (code yihishe) muri autoencoder?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Niyihe ntego ya 'icupa' (code yihishe) muri autoencoder?
    • Guhatira umuyoboro muburyo bworoshye bufata imiterere yingenzi
    • Kongera ibipimo byinshi no kongera ubushobozi bwicyitegererezo
    • Kubika ibirango byo kugenzura ibyiciro
    • Kugabanya imyitozo kugirango ituze neza
  2. Kuki autoencoders ifatwa nkigenzurwa?
    • Bakenera imibare nini yanditseho imibare
    • Intego y'amahugurwa niyinjiza ubwayo, ntabwo rero ibirango byo hanze bikenewe
    • Umuntu aranga buri kwiyubaka n'intoki
    • Bakora gusa kumashusho yabanjirije
  3. Nigute gutandukanya autoencoder itandukanye nibisanzwe?
    • Ikuraho decoder rwose
    • Ikora gusa hamwe namakuru yamakuru
    • Yatojwe kongera kubaka ibintu bisukuye bivuye muri verisiyo yangiritse
    • Ntabwo bisaba ko habaho icyuho
  4. Niki gituma autoencoder itandukanye (VAE) ibasha gutanga amakuru mashya?
    • Ifata mu mutwe ingero zose zamahugurwa
    • Ntabwo ikoresha decoder
    • Isimbuza imiyoboro yimitsi nibiti byicyemezo
    • Yiga neza, bishoboka ko byihishe umwanya ushobora kwigana
  5. Umurongo wa autoencoder yatojwe hamwe na kare-kwaduka ikosa risa nubuhanga bwa kera?
    • Isesengura ryibanze ryibanze (PCA)
    • K-bisobanura guhuriza hamwe
    • Ibiti
    • Gusubira inyuma
  6. Kuki autoencoders ikwiranye no gutahura anomaly?
    • Bandika ibintu bidasanzwe mugihe cy'amahugurwa
    • Bongeye kubaka imiterere isanzwe ariko berekana amakosa manini kubidasanzwe, bitagaragara
    • Birengagije amakosa yo kwiyubaka rwose
    • Basohoka gusa ibyiciro bibiri
  7. Which reconstruction loss does the guide associate with binary-pixel data?
    • Igihombo cya gatatu
    • Temporal-difference error
    • Cross-entropy
    • Cosine annealing loss
  8. Why can nonlinear activations make an autoencoder more expressive than a linear one?
    • They force the latent code to equal the input size
    • They remove the decoder from the model
    • They guarantee zero reconstruction error on unseen data
    • They allow it to learn nonlinear or curved structures in the data