Reinforcement Learning CNN

Refer to:
1.
Paper:
https://arxiv.org/pdf/1406.6247.pdf on June 2014
Source:
https://github.com/fatalfeel/Recurrent_Model_Visual_Attention

2.
Paper on February 2018:

https://www.researchgate.net/publication/322969740_An_Aircraft_Detection_Framework_Based_on_Reinforcement_Learning_and_Convolutional_Neural_Networks_in_Remote_Sensing_Images/fulltext/5a7a56d6aca2722e4df4fce1/An-Aircraft-Detection-Framework-Based-on-Reinforcement-Learning-and-Convolutional-Neural-Networks-in-Remote-Sensing-Images.pdf

IoU actions:
1. top-left
2. top-right
3. bottom-left
4. bottom-right
5. middle
6. end (entire image)

This paper can use LunarLander which have 4 actions to implement
https://github.com/fatalfeel/PPO_Pytorch/blob/master/PPO_LunarLander_train.py

3.
Reinforcement learning and convolutional neural network
https://www.ericsson.com/en/blog/2018/9/demystifying-reinforcement-learning-and-convolutional-neural-network

4.
Paper:
https://arxiv.org/pdf/1701.02392.pdf on January 2017
Source:
https://github.com/tanmayshankar/RCNN_MDP
Video:
https://www.bilibili.com/video/av78766875

Basic Theory:
(a.) LSTM on December 1997
https://www.researchgate.net/profile/Sepp_Hochreiter/publication/13853244_Long_Short-term_Memory/links/5700e75608aea6b7746a0624/Long-Short-term-Memory.pdf
(b.) RNN & LSTM
https://www.youtube.com/watch?v=xCGidAeyS4M
https://www.youtube.com/watch?v=rTqmWlnwz_0
(c.) LSTM backpropagation
https://medium.com/@ZhengLungWu/lstm%E6%AD%A3%E5%8F%8D%E5%90%91%E5%82%B3%E6%92%AD%E6%8E%A8%E5%B0%8E-3732055b15e9
(d.) There are layer1, layer2 LSTM hand made examples which are important
https://www.mediafire.com/file/qkqpymgpppi9ark/torch_lstm.py

https://www.mediafire.com/view/gsh76xxf05569ec/lstm_1l.png (LSTM 1 layer map) 

https://www.mediafire.com/view/4te5lu9ugoxkav3/lstm_2l.jpg (LSTM 2 layers map)
(e.) Pytorch LSTM
https://pytorch.org/docs/stable/generated/torch.nn.LSTM.html
(f.) POMDP
https://people.eecs.berkeley.edu/~pabbeel/cs287-fa13/slides/pomdps.pdf
(g.) Recurrent Model of Visual Attention
http://torch.ch/blog/2015/09/21/rmva.html
https://www.cnblogs.com/wangxiaocvpr/p/5537454.html

https://www.youtube.com/watch?v=kcLuHWrn-t0

(h.) NLL-Loss & CrossEntropyLoss
https://www.mediafire.com/file/n0ron4zkqhu5vgt/nll_crossEntropy.py

(i.) affine grid & grid sample source

https://www.mediafire.com/file/l2sawg5hepxucn0/affinegrid_gridsample.zip

Forge a voice in RNN
https://www.youtube.com/watch?time_continue=332&v=9Yq67CjDqvw&feature=emb_title

Tesla told in New York Herald: I prefer to be remembered as the inventor who succeeded in abolishing war. That will be my highest pride.
http://www.teslacollection.com/tesla_articles/1898/new_york_herald/f_l_christman/tesla_declares_he_will_abolish_war (in middle section)

Albert Einstein: The release of atom power has changed everything except our way of thinking... the solution to this problem lies in the heart of mankind. If only I had known, I should have become a watchmaker.
https://atomictrauma.wordpress.com/the-scientists/albert-einstein