sji

  • 홈
  • 태그
  • 방명록

Interview 1

ML General 잡질문/답변 (기술면접)

기존 글에도 정리를 해왔지만, 기술 면접 준비하면서 정리 했던 것들 몇 가지 보충해서 정리해봤다. 틀린 내용이 있을 수 있으니 의심하면서 보길... Backpropagation 계산법 (이상한 곳에서는 간단한 계산 시킬 수도...) 로컬 그래디언트에 업스트림 그래디언트를 곱해줌으로써 구하는 것을 기본 원칙으로 계산 하면 된다 Regression vs. Classification 확률적인 관점에서 설명할 줄 알면 든든... 가정하는 분포가 뭔지 (이산? 연속? / 베르누이? 가우시안?) 가정 분포에 따라 Loss는 어떻게 달라지는지... Sigmoid 쓰는 이유? In order to map predicted value to probability a는 sigmoid의 기울기를 결정, b는 중심 위치를 결정..

DL&ML/concept 2021.05.14
1
더보기
  • 분류 전체보기 (87)
    • intro. (0)
    • DL&ML (38)
      • concept (19)
      • papers (9)
      • code.data.tips (10)
    • http&server (7)
    • dist. system (6)
    • docker (5)
    • algorithm (24)
    • design pattern (1)
    • architecture (1)
    • trouble shooting (3)
    • better way (1)
    • problems (0)

Tag

Docker, summarization, React Native, tree, docker-compose, stack, string, seq2seq, Graph, Zookeeper, flask, queue, Dynamic Programming, gaussian mixture, CrossEntropy, gunicorn, nginx, dp, MLE, Likelihood,

최근글과 인기글

  • 최근글
  • 인기글
250x250

Calendar

«   2026/01   »
일 월 화 수 목 금 토
1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31

방문자수Total

  • Today :
  • Yesterday :

Designed by Tistory.

티스토리툴바