Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[id] cs-229-deep-learning #154
base: master
Are you sure you want to change the base?
[id] cs-229-deep-learning #154
Changes from 4 commits
941389e
3712acc
152376a
a789bcd
2808c50
File filter
Filter by extension
Conversations
Jump to
There are no files selected for viewing
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Tahap 3: Lakukan perhitungan backpropagate dengan nilai loss untuk mendapatkan nilai gradien.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Tahap 4: Gunakan gradien untuk mengubah nilai weights.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Droput - Dropout adalah sebuah teknik yang digunakan untuk mencegah overfitting pada saraf tiruan dengan memutus unit yang terdapat pada sebuah neural network. Dalam praktiknya, neuron dikurangi dengan probabilitas p atau dipertahankan dengan probabilitas 1-p
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Convolutional Neural Network
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Gun ini udah gw perbaikin, bisa kita close?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Bisa Bang kalo udah gak ada tambahan/perbaikan lagi.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Normalisasi batch - Normalisasi batch adalah sebuah langkah dari hiperparameter γ,β untuk menormalisasi batch {xi}. Dengan mendefinisikan μB,σ2B sebagai nilai rata-rata dan variansi dari batch yang ingin kita normalisasi, hal tersebut dapat dilakukan dengan cara:
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Normalisasi batch biasa ditempatkan setelah sebuah layer yang sepenuhnya terhubung/konvolusi dan sebelum sebuah layer non-linear yang bertujuan untuk memungkinkannya penggunaan nilai learning rate yang lebih tinggi dan mengurangi ketergantungan kuat pada nilai inisialisasi parameter.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Recurrent Neural Networks
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Tipe-tipe gerbang - Di bawah ini merupakan beberapa jenis gerbang yang biasa ditemui pada recurrent neural network:
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LSTM - Long short-term memory (LSTM) network adalah sebuah tipe model dari RNN yang mencegah permasalahan gradien vanishing (nilai gradien menjadi 0) dengan menambahkan gerbang 'lupa'.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Tujuan dari reinforcement learning adalah sebagai sebuah agen (contoh: robot) yang dapat belajar untuk menyesuaikan diri terhadap lingkungan sekelilingnya.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Markov decision processes - Markov decision process (MDP) adalah sebuah 5-tuple(S,A,{Psa},γ,R) di mana:
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
S adalah bagian dari tahap-tahap
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
A adalah bagian dari aksi-aksi
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
{Psa} adalah transisi probabilitas dari satu tahap ke tahap lain untuk s∈S and a∈A
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
R:S×A⟶R or R:S⟶R adalah fungsi reward (hadiah) yang algoritma ingin memaksimalkan nilai keluarannya
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Catatan: kami menemukan bahwa kebijakan optimal π∗ untuk state yang diberikan seperti:
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Algoritma value iteration - Algoritma value iteration memiliki dua tahap:
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Kita menginialisasi value:
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Masa melakukan aksi a pada tahap s