이전에 다뤘던 포스트 중에 ANN을 사용해서 Churn prediction을 했던 내용이 있다. 간단히 말해 개인에 대한 정보를 바탕으로 이 사람이 credit이 있는지 없는지 여부를 판단해주는 예제였다. 그때 사용했던 데이터를 보통 데이터 필드가 numerical variable도 있고, categorical variable도 있었다. 물론 categorical variable을 ML이나 Deep Learning에서 다루기 위해서는 뭔가 의미있는 정보로 변화시켜주는 일련의 Encoding 과정이 필요했고, 그 때 기억으로는 Scikit-learn에서 제공하는 LabelEncoder와 OneHotEncoder를 사용해서 데이터를 Binary 처리를 하고, 학습에 반영했다. 아마 이렇게 처리하는 방식이 C..
(OpenAI Spinning Up 글을 개인적으로 정리했습니다. 원본) Extra Material — Spinning Up documentation Docs » Extra Material Edit on GitHub © Copyright 2018, OpenAI. Revision 97c8c342. Built with Sphinx using a theme provided by Read the Docs. spinningup.openai.com 이번 글에서는 finite-horizon undiscounted return 상태에서 다음 식을 증명하고자 한다. $$ \nabla_{\theta} J(\pi_{\theta}) = E_{\tau \sim \pi_{\theta}} \Big[ \sum_{t=0}^{T} \b..
(이 글은 OpenAI Spinning Up의 글을 개인적으로 정리한 내용입니다. 원본) Extra Material — Spinning Up documentation Docs » Extra Material Edit on GitHub © Copyright 2018, OpenAI. Revision 97c8c342. Built with Sphinx using a theme provided by Read the Docs. spinningup.openai.com 이번 글에서는 action이 이전에 얻은 reward에 reinforce되서는 안된다는 것을 증명하고 한다. 먼저 simplest policy gradient에서의 식 중 \(R(\tau)\)를 전개해보면 다음과 같다. $$ \begin{align} \n..
( 본 글은 OpenAI Spinning Up을 개인적으로 정리한 글입니다. 원본) Part 3: Intro to Policy Optimization — Spinning Up documentation In this section, we’ll discuss the mathematical foundations of policy optimization algorithms, and connect the material to sample code. We will cover three key results in the theory of policy gradients: In the end, we’ll tie those results together and desc spinningup.openai.com 이번 글에서는..

(본 글은 OpenAI Spinning Up을 개인적으로 정리한 글입니다. 원본) Part 2: Kinds of RL Algorithms — Spinning Up documentation We’ll start this section with a disclaimer: it’s really quite hard to draw an accurate, all-encompassing taxonomy of algorithms in the modern RL space, because the modularity of algorithms is not well-represented by a tree structure. Also, to make somethin spinningup.openai.com RL Algorithm의 ..

(본 글은 OpenAI Spinning Up 글을 개인적으로 정리한 글입니다. 원본) Part 1: Key Concepts in RL — Spinning Up documentation A state is a complete description of the state of the world. There is no information about the world which is hidden from the state. An observation is a partial description of a state, which may omit information. In deep RL, we almost always represent stat spinningup.openai.com RL을 다루면 가장 많이 나오..

회사 지원을 받아서 미국으로 나와있다. 미국에 나오자마자 살려고 고민했던게 바로 Google에서 출시한 Coral Dev Board였는데, 이렇게 수령했다. 대충 가격은 150정도에 Tax 붙고 배송비까지 적용하니까 180불정도 했던거 같다. 이게 눈에 띄는 이유는 Google에서 만든 NPU인 TPU가 on-board 형식으로 embedded된 것이다. 지금까지 나온 형태는 Chip내에 내장되어 있던가, 아니면 Movidius에서 Neural Computing Stick 처럼 USB를 통해 처리를 지원받는 형식이었는데, 이건 위처럼 보드 타입도 있고, usb-c type으로 연결해 쓰는 타입도 있다. 나름 구글에서는 edge computing, fast ML을 지향하면 출시한 것이긴 한데, 예제좀 돌려..
- Total
- Today
- Yesterday
- SketchFlow
- Windows Phone 7
- Expression Blend 4
- Off-policy
- TensorFlow Lite
- 강화학습
- processing
- End-To-End
- Kinect
- Kinect for windows
- ColorStream
- arduino
- 파이썬
- Offline RL
- DepthStream
- windows 8
- 딥러닝
- reward
- Variance
- PowerPoint
- Gan
- dynamic programming
- Kinect SDK
- bias
- Policy Gradient
- 인공지능
- RL
- Distribution
- Pipeline
- 한빛미디어
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 |