Pembelajaran Mesin (Machine Learning) Bagian 1

Merupakan catatan pribadi dari pra-kuliah materi perdana (pertemuan kedua)

Basic of Neural Network (NN) :

  1. Used to predict missing data
  2. Uses a compilation of data (list) as the reference for prediction
  3. Sigmoid(x) is used to range between 0 and 1
  4. Weight starts at random, adapts with each correction
  5. Square error cost for accuracy is needed : cost (b) = (b-r)^2
  6. Utilize slope for better accuracy where b = b*lrn*slope
  7. There are 3 calculation methods for slope: numerical, algebraically, calculus (derivative)
[Q1] Can computer become hyper-intelligent enough to threaten humanity?

Komentar

Postingan populer dari blog ini

[HOAX] Pesan Juru Kunci

Daftar Enzim Pencernaan, Letak dan Fungsi

Typeform : Membuat Formulir dengan Mudah!