- Deep Q Learning for Video Games - The Math of Intelligence #9

Deep Q Learning for Video Games - The Math of Intelligence #9

We're going to replicate DeepMind's Deep Q Learning algorithm for Super Mario Bros! This bot will be able to play a bunch of different video games by using reinforcement learning. This is the first video in this series that uses libraries (Keras & Gym) because if it didn't, the code would be ...
We're going to replicate DeepMind's Deep Q Learning algorithm for Super Mario Bros! This bot will be able to play a bunch of different video games by using reinforcement learning. This is the first video in this series that uses libraries (Keras & Gym) because if it didn't, the code would be way too long for a short video. I'll make a longer, in-depth version without libraries soon.

Code for this video:
https://github.com/llSourcell/deep_q_learning

Please Subscribe! And like. And comment. That's what keeps me going.

More learning resources:
https://medium.com/emergent-future/simple-reinforcement-learning-with-tensorflow-part-0-q-learning-with-tables-and-neural-networks-d195264329d0
http://pytorch.org/tutorials/intermediate/reinforcement_q_learning.html
http://neuro.cs.ut.ee/demystifying-deep-reinforcement-learning/
http://karpathy.github.io/2016/05/31/rl/
https://yanpanlau.github.io/2016/07/10/FlappyBird-Keras.html
https://keon.io/deep-q-learning/
http://www0.cs.ucl.ac.uk/staff/d.silver/web/Resources_files/deep_rl.pdf
http://mnemstudio.org/path-finding-q-learning-tutorial.htm

Join us in the Wizards Slack channel:
http://wizards.herokuapp.com/

And please support me on Patreon:
https://www.patreon.com/user?u=3191693
Follow me:
Twitter: https://twitter.com/sirajraval
Facebook: https://www.facebook.com/sirajology Instagram: https://www.instagram.com/sirajraval/
Signup for my newsletter for exciting updates in the field of AI:
https://goo.gl/FZzJ5w
Hit the Join button above to sign up to become a member of my channel for access to exclusive content!

#deep q learning atari #deep q-learning #deep q network #deep q learning #deep q #deep q-learning algorithm #deep q-learning tutorial #deep learning game #q learning #deep q learning python #deep q learning tutorial #q-learning #deep reinforcement learning #deep q-learning with recurrent neural networks
it's damn hilarious. Keep going! - Deep Q Learning for Video Games - The Math of Intelligence #9

it's damn hilarious. Keep going!

Deep Q Learning for Video Games - The Math of Intelligence #9
2017年08月12日
00:00:53 - 00:09:47
That picture  of Siraj petting a lama at  should be the cover of his mixtape, unsupervised learner great video Siraj! - Deep Q Learning for Video Games - The Math of Intelligence #9

That picture of Siraj petting a lama at should be the cover of his mixtape, unsupervised learner great video Siraj!

Deep Q Learning for Video Games - The Math of Intelligence #9
2017年08月12日
00:01:19 - 00:09:47
atyou say the more in the fuure the reward is - more are we uncertain of it? i didn't get it-can you explain with an example ? - Deep Q Learning for Video Games - The Math of Intelligence #9

atyou say the more in the fuure the reward is - more are we uncertain of it? i didn't get it-can you explain with an example ?

Deep Q Learning for Video Games - The Math of Intelligence #9
2017年08月12日
00:05:15 - 00:09:47
Well I don't think the pooling layer is used to get insensitive about the locations of the objects in an image. The convolutional layer can already do that since the convolutional operation is actually a pixel window going from location to location until all locations are considered under the set stride. The pooling layer is used to semantically merge similar features into one, like in the max pooling example used in this video, you can see the image is partitioned into 4 parts and in each part, the max number is preserved. The max number can semantically represent a feature in that region. It's more like image compression but we have preserved the key features of this object in this image. Feeding this pooled image into the neural net could be more efficient. - Deep Q Learning for Video Games - The Math of Intelligence #9

Well I don't think the pooling layer is used to get insensitive about the locations of the objects in an image. The convolutional layer can already do that since the convolutional operation is actually a pixel window going from location to location until all locations are considered under the set stride. The pooling layer is used to semantically merge similar features into one, like in the max pooling example used in this video, you can see the image is partitioned into 4 parts and in each part, the max number is preserved. The max number can semantically represent a feature in that region. It's more like image compression but we have preserved the key features of this object in this image. Feeding this pooled image into the neural net could be more efficient.

Deep Q Learning for Video Games - The Math of Intelligence #9
2017年08月12日
00:07:46 - 00:09:47
At  whats the input_shape supposed to be ?? the challenge code and what you show are different ...... - Deep Q Learning for Video Games - The Math of Intelligence #9

At whats the input_shape supposed to be ?? the challenge code and what you show are different ......

Deep Q Learning for Video Games - The Math of Intelligence #9
2017年08月12日
00:08:08 - 00:09:47
I understand that a convolutional neural network can be used to simplify the state from an array of pixels to a smaller collection of values, but how does the algorithm use a deep network to approximate the Q-function? - Deep Q Learning for Video Games - The Math of Intelligence #9

I understand that a convolutional neural network can be used to simplify the state from an array of pixels to a smaller collection of values, but how does the algorithm use a deep network to approximate the Q-function?

Deep Q Learning for Video Games - The Math of Intelligence #9
2017年08月12日
00:08:19 - 00:09:47
and it's only  long? Autolike from me :) - Deep Q Learning for Video Games - The Math of Intelligence #9

and it's only long? Autolike from me :)

Deep Q Learning for Video Games - The Math of Intelligence #9
2017年08月12日
00:09:46 - 00:09:47
Siraj Raval

Siraj Raval

🎉 710,000 人達成! 🎉

チャンネル登録 RSS
Hello World, it's Siraj! I'm a technologist on a mission to spread data literacy. Artificial Intelligence, Mathematics, Science, Technology, I simplify these topics to help you understand how they work. Using this knowledge you can build wealth and live a happier, more meaningful life. I live to...
Hello World, it's Siraj! I'm a technologist on a mission to spread data literacy. Artificial Intelligence, Mathematics, Science, Technology, I simplify these topics to help you understand how they work. Using this knowledge you can build wealth and live a happier, more meaningful life. I live to serve this community. We are the fastest growing AI community in the world! Co-Founder of Sage Health (www.sage-health.org)

Twitter: http://www.twitter.com/sirajraval
Instagram: https://instagram.com/sirajraval/
Facebook: https://www.facebook.com/sirajology/

If you found my videos useful, I'd love your support on Patreon :) https://www.patreon.com/user?ty=h&u=3191693

Some Research - https://drive.google.com/file/d/0BwUv84lNDk72Q1gzaXgwR2U3U2NWVlZSOFk4amZIRmV1QXI0/view

In the event of my demise, you must finish what I've started here.

Timetable

動画タイムテーブル

動画数:372件

At  we see it costs ~19$ to upload a single video! (ETH goes ~4000$ on Dec 28,2021 and fees are 0.004782ETH). Am I missing something? - Build a Social Media Dapp with Polygon

At we see it costs ~19$ to upload a single video! (ETH goes ~4000$ on Dec 28,2021 and fees are 0.004782ETH). Am I missing something?

Build a Social Media Dapp with Polygon
2021年12月28日
00:02:17 - 00:49:20
whats the video man whats wrong with u - Building a Health DAO with GitHub CoPilot (AlphaCare: Episode 5)

whats the video man whats wrong with u

Building a Health DAO with GitHub CoPilot (AlphaCare: Episode 5)
2021年11月09日
00:00:47 - 00:44:37
SHIB to the moon - Building a Health DAO with GitHub CoPilot (AlphaCare: Episode 5)

SHIB to the moon

Building a Health DAO with GitHub CoPilot (AlphaCare: Episode 5)
2021年11月09日
00:24:35 - 00:44:37
Wow! My code at , 3:18 Thank you so much, Siraj. I am honored. - Multiomics Data for Cancer Diagnosis (AlphaCare: Episode 3)

Wow! My code at , 3:18 Thank you so much, Siraj. I am honored.

Multiomics Data for Cancer Diagnosis (AlphaCare: Episode 3)
2021年10月16日
00:01:42 - 00:10:57
I'm afraid that I don't understand this diagram indicating that proteins go throuh metabolism and end up with biochemicals like carbohydrates. There are some mechanisms that carbohydrates can be switched into amino acids and vice versa,  but there are also amino acids that human body can not produce. Plus how lipids can be made out of proteins...? - Multiomics Data for Cancer Diagnosis (AlphaCare: Episode 3)

I'm afraid that I don't understand this diagram indicating that proteins go throuh metabolism and end up with biochemicals like carbohydrates. There are some mechanisms that carbohydrates can be switched into amino acids and vice versa, but there are also amino acids that human body can not produce. Plus how lipids can be made out of proteins...?

Multiomics Data for Cancer Diagnosis (AlphaCare: Episode 3)
2021年10月16日
00:03:37 - 00:10:57
I appreciated the kind of effort you have devoted for this  min video .keep going siraj big fan from india. - Perceiver for Cardiac Video Data Classification (AlphaCare: Episode 2)

I appreciated the kind of effort you have devoted for this min video .keep going siraj big fan from india.

Perceiver for Cardiac Video Data Classification (AlphaCare: Episode 2)
2021年10月02日
00:11:40 - 00:11:41
Did anyone else have to rewind and double check what he said at ? Is that actually a real thing? Surely not - Convolutional Networks for Heart Disease Prediction (AlphaCare: Episode 1)

Did anyone else have to rewind and double check what he said at ? Is that actually a real thing? Surely not

Convolutional Networks for Heart Disease Prediction (AlphaCare: Episode 1)
2021年08月20日
00:01:57 - 00:08:57
at this point it's clear to me that you don't know what you're talking about. - Yolo V5 Snowboarding LIVE

at this point it's clear to me that you don't know what you're talking about.

Yolo V5 Snowboarding LIVE
2020年12月28日
00:23:00 - 01:00:35
Oh, Is that a social network site for coders? - Yolo V5 Snowboarding LIVE

Oh, Is that a social network site for coders?

Yolo V5 Snowboarding LIVE
2020年12月28日
00:23:03 - 01:00:35
I'm at  and he just said that he got docker, tensorflow and cuda running like wtf? You can't have CUDA without Nvidia GPUs and docker already addressed that they don't support Apple M1 chip yet.Wtf. He has no idea what the fuck he is saying I can guarantee. - Deep Learning on Apple M1 Silicon LIVE

I'm at and he just said that he got docker, tensorflow and cuda running like wtf? You can't have CUDA without Nvidia GPUs and docker already addressed that they don't support Apple M1 chip yet.Wtf. He has no idea what the fuck he is saying I can guarantee.

Deep Learning on Apple M1 Silicon LIVE
2020年12月16日
00:00:49 - 01:06:21
Sorry, that's just wrong. Apple M1 is ARM-based. Read the first sentence on https://en.wikipedia.org/wiki/Apple_M1. And it is by no means Apple's first CPU, it's Apple's first PC CPU. There are well-established toolchains for ARM, and many open source projects just need to be recompiled to run, they don't need to be "ported". - Deep Learning on Apple M1 Silicon LIVE

Sorry, that's just wrong. Apple M1 is ARM-based. Read the first sentence on https://en.wikipedia.org/wiki/Apple_M1. And it is by no means Apple's first CPU, it's Apple's first PC CPU. There are well-established toolchains for ARM, and many open source projects just need to be recompiled to run, they don't need to be "ported".

Deep Learning on Apple M1 Silicon LIVE
2020年12月16日
00:07:20 - 01:06:21
all good ? - Let's Build Machine Learning...in RUST? LIVE

all good ?

Let's Build Machine Learning...in RUST? LIVE
2020年11月28日
00:10:22 - 01:02:11
- 25:10 -- getting some real Trump vibes. - Let's Build Machine Learning...in RUST? LIVE

- 25:10 -- getting some real Trump vibes.

Let's Build Machine Learning...in RUST? LIVE
2020年11月28日
00:25:06 - 01:02:11
Use the sandbox on the rust website? - Let's Build Machine Learning...in RUST? LIVE

Use the sandbox on the rust website?

Let's Build Machine Learning...in RUST? LIVE
2020年11月28日
00:34:00 - 01:02:11
It starts at  btw - Let's Build Machine Learning...in RUST? LIVE

It starts at btw

Let's Build Machine Learning...in RUST? LIVE
2020年11月28日
00:35:01 - 01:02:11
Amazing 😃 - Let's Build a Quantum Classifier! LIVE

Amazing 😃

Let's Build a Quantum Classifier! LIVE
2020年11月21日
00:06:07 - 01:13:20
*Accessing Parallal Timelines ! ! ! (****)* - Let's Build a Quantum Classifier! LIVE

*Accessing Parallal Timelines ! ! ! (****)*

Let's Build a Quantum Classifier! LIVE
2020年11月21日
00:30:59 - 01:13:20
1/x is totally differentiable. It's just x^(-1), multiply by power and subtract one from it gives -(x^(-2)) i.e. -1/x^2. Surely you can't do anything advanced like back propagation without understanding that. - Let's Build a Quantum Classifier! LIVE

1/x is totally differentiable. It's just x^(-1), multiply by power and subtract one from it gives -(x^(-2)) i.e. -1/x^2. Surely you can't do anything advanced like back propagation without understanding that.

Let's Build a Quantum Classifier! LIVE
2020年11月21日
01:04:15 - 01:13:20
AM PST/ PM IST! - Let's Build Data Structures & Algorithms! LIVE

AM PST/ PM IST!

Let's Build Data Structures & Algorithms! LIVE
2020年11月07日
00:09:30 - 00:54:05