퀴즈 · 4 questions

🏗️ Architecture: Vision, Sequence, Attention

CNN, recurrence, attention, transformer

Level 0Curious

0 XP0/73 lessons0/11 achievements

0/120 XP to next level120 XP to go0% complete

01What's the most fundamental advantage of a CNN over an MLP for image classification?

02Why are residual connections so universal in modern architectures (ResNet, Transformer, U-Net, etc.)?

03What does scaled dot-product attention compute?

04What's the difference between encoder-only (BERT) and decoder-only (GPT) transformer attention?

이 페이지에서 버그를 발견하셨거나 피드백이 있으세요?문제 신고

로그인 — 댓글을 남기려면 로그인해 주세요.

아직 댓글이 없어요. 첫 댓글을 남겨보세요.