Ikibazo gihuza ibibazo · Hagati Urwego

Ubushakashatsi bwa Neural Ubwubatsi

Ongera usuzume NAS ishakisha, algorithms, igiciro cyo gushakisha hakiri kare, DARTS kuruhuka, kugabana ibiro, hamwe no gusuzuma caveats. Reba ibitekerezo n'ibipimo.

Inzira zijyanye no kuyoboraGushakisha Ubwubatsi bwa Neural
Ikibazo 1 Bya 8

Ni ubuhe buryo bwa Neural Architecture Shakisha bwikora?

Reba buri kibazo8 ibibazo

Ibisubizo nibisobanuro bigaragara mugihe ufata ikibazo.

  1. Ni ubuhe buryo bwa Neural Architecture Shakisha bwikora?
    • Igishushanyo cyimiterere yimiyoboro
    • Kwandika amakuru yamahugurwa
    • Gukonjesha seriveri ya GPU
    • Kwandika inyandiko zerekana icyitegererezo
  2. Nibihe bice bitatu bisobanura uburyo bwa NAS?
    • Dataset, optimizer, hamwe nabata ishuri
    • Umwanya wo gushakisha, ingamba zo gushakisha, hamwe no kugereranya imikorere
    • Encoder, decoder, hamwe no kwitondera
    • Amahugurwa, kwemeza, hamwe nibizamini
  3. What accelerator scale did the original reinforcement-learning NAS paper report for its large search?
    • One GPU for every NAS method
    • A fixed cost of exactly 1,000 GPU-days for every search
    • An experiment using 800 GPUs in that study
    • No accelerator use during architecture search
  4. Ni ubuhe buryo bw'ingenzi DARTS ikoresha kugirango ishakisha rikore neza?
    • Itoza kurugero rumwe
    • Ikuraho ibice byose byihishe
    • Ikoresha gushakisha gusa
    • Iruhura imyubakire itandukanye muburyo bukomeza neza na gradients
  5. Niki 'supernet' mugusangira ibiro NAS?
    • Umuyoboro umwe urenze urugero urimo abakandida bose bubaka, uburemere bwabo busangiwe
    • Umuyoboro ufite miliyari y'ibipimo
    • Umuyoboro ukora kuri mudasobwa zidasanzwe gusa
    • Gukoporora kopi yicyitegererezo
  6. What did the HURRICANE hardware-aware NAS study evaluate?
    • One architecture that is guaranteed optimal on every device
    • Tailored architectures for three hardware types under matched latency constraints
    • A language model trained without any hardware measurements
    • A method that removes latency from architecture evaluation
  7. What is one limitation of weight-sharing NAS supernets?
    • They remove all architecture choices
    • They guarantee the globally best architecture
    • The shared weights can bias architecture rankings relative to training each candidate independently
    • They do not use neural networks
  8. What does a NAS evaluation need to report for a fair comparison?
    • Only the best test score
    • The GPU brand alone
    • The model name without settings
    • Search space, compute budget, validation protocol, and final retraining setup