Ujian A/B untuk Kuiz Model ML

Jawapan dan penjelasan muncul semasa anda mengambil kuiz.

1. Mengapakah pasukan A/B menguji model dalam pengeluaran dan bukannya mempercayai metrik luar talian?
   • Model dengan ketepatan luar talian yang lebih baik masih boleh menjejaskan metrik perniagaan sebenar
   • Metrik luar talian adalah haram
   • Pengeluaran lebih murah daripada ujian luar talian
   • Data luar talian tidak wujud
2. Apakah peranan yang dimainkan oleh tugasan rawak dalam ujian A/B?
   • Ia menjadikan ujian berjalan lebih cepat
   • It supports balance between groups in expectation, subject to design and implementation assumptions
   • Ia memadam pengguna berat sebelah
   • Ia mengurangkan bilangan model yang diperlukan
3. In a model A/B test, what usually happens in the treatment group?
   • The candidate scores requests but all outputs are withheld
   • Every user is assigned to both models for the same decision
   • The candidate model serves user-facing decisions for the randomly assigned group
   • The control model is removed before any comparison
4. Under a null hypothesis and stated test assumptions, how is a p-value interpreted?
   • As the probability that the null hypothesis is true
   • As the probability that the treatment caused the observed result
   • As the probability that the experiment will replicate
   • As the probability of data at least this extreme if the null hypothesis holds
5. How can a multi-armed bandit change an experiment while it runs?
   • It can allocate more traffic to variants with better observed outcomes and less to others
   • It keeps every variant at a fixed equal allocation throughout
   • It removes randomization from every assignment
   • It guarantees the best variant is identified without uncertainty
6. Apakah tujuan analisis kuasa sebelum ujian A/B?
   • Untuk memilih logo untuk percubaan
   • Untuk menganggarkan bilangan pengguna yang diperlukan untuk mengesan kesan sebenar dengan pasti
   • Untuk menyulitkan trafik
   • Untuk memadam kumpulan kawalan
7. What does a sample-ratio mismatch (SRM) signal during an A/B test?
   • The treatment model is necessarily better
   • The experiment has sufficient statistical power
   • Observed assignment counts differ materially from the configured allocation, which can indicate a data or randomization problem
   • The primary metric improved causally
8. Why include guardrail metrics alongside an experiment’s primary success metric?
   • To guarantee the candidate wins on every metric
   • To replace the primary metric after results are observed
   • To avoid defining a decision rule before the test
   • To detect meaningful harms in other outcomes even when the primary metric improves