- 35
- 586 584
Sebastian Schuchmann
Приєднався 17 чер 2017
A.I. and Machine Learning for everybody! Sometimes I am teaching machines how to play games via Reinforcement Learning in Unity3D and sometimes I am exploring the deep history of A.I.
The tools I use: Unity3D with ML-Agents.
The tools I use: Unity3D with ML-Agents.
I will showcase/fix your Unity project for FREE!
Twitter: SebastianSchuc7
Email also works: schuchmannsebastian@gmail.com
Send me your Unity Project to have it featured or if you want me to take a look at it. It doesn't have to be AI/ML-Agents - just a Unity Project. It will take a few weeks for me to handle the projects but you can expect the video this year!
Support me on Patreon: www.patreon.com/user?u=25285137
ML-Agents Discord Channel: discord.gg/wDPWsQT
Email also works: schuchmannsebastian@gmail.com
Send me your Unity Project to have it featured or if you want me to take a look at it. It doesn't have to be AI/ML-Agents - just a Unity Project. It will take a few weeks for me to handle the projects but you can expect the video this year!
Support me on Patreon: www.patreon.com/user?u=25285137
ML-Agents Discord Channel: discord.gg/wDPWsQT
Переглядів: 1 170
Відео
Codex, the model that has learned to code
Переглядів 4,3 тис.2 роки тому
It's been a long time since a proper video but this is so big I just couldn't let it slide. Codex Challenge: challenge.openai.com/ Support me on Patreon: www.patreon.com/user?u=25285137 ML-Agents Discord Channel: discord.gg/wDPWsQT Keep in touch: SebastianSchuc7
AI Tier List (Alpha Go, GPT-3, Dall-E ...)
Переглядів 1,7 тис.3 роки тому
Ranking different AI's throughout history because it's fun, why not. Don't worry, I am still working on ML-Agents stuff, but time is currently tight. 00:15: Alpha Go 01:10: IBM's Deep Blue 01:57: GPT-3 02:25: Eliza Chatbot 03:00: Arthur Samuels Checkers 03:48: Shakey 04:24: Dall-E 05:01: Mark 1 Perceptron 05:23: Microsofts Tay Bot Support me on Patreon: www.patreon.com/user?u=25285137 ML-Agents...
A GPT-3 for Images? Dall-E is the most impressive AI ever created!
Переглядів 36 тис.3 роки тому
DALL·E / Dall-E is a model based on GPT-3 but for generating images. In the realm of Machine Learning or AI, this has to one of the most impressive models ever released. OpenAI again pushes the boundaries of what's possible. Support me on Patreon: www.patreon.com/user?u=25285137 ML-Agents Discord Channel: discord.gg/wDPWsQT Keep in touch: SebastianSchuc7 Original Article: openai.com...
PyTorch Beginner Tutorial - Training an Image Classification Model and putting it online!
Переглядів 32 тис.3 роки тому
In this tutorial we'll be training a Deep Learning Model with PyTorch 1.7 and FastAI in a beginner-friendly way. I will explain how to use Machine Learning and the basics about how it works. At the end you will have a public website up and running where your Model is deployed using Flask, Python and Docker. Sorry for some of the audio issues! *Links related to Tutorial*: Github Link: github.com...
Creating Mario Party in Augmented Reality
Переглядів 1,5 тис.3 роки тому
So proud to share a work-related project I was part of. As a team of two we put our heart into developing a Mario Party-like game for AR, called AR Party. Much love goes out for any support! :) Kickstarter: www.kickstarter.com/projects/ditached/ar-party?ref=cdahf4 This may not be my usual kind of video but I am really proud of what we have accomplished and believed that some of you might enjoy ...
ML-Agents 1.0+ Creating a Mario Kart like AI
Переглядів 35 тис.3 роки тому
A Unity ML-Agents tutorial for creating a machine learning AI that drives similar to Mario Kart. Keep in touch: SebastianSchuc7 Support me on Patreon: www.patreon.com/user?u=25285137 ML-Agents Discord Channel: discord.gg/wDPWsQT Links: Github Repository (Project): github.com/Sebastian-Schuchmann/AI-Racing-Karts Unity Download: unity3d.com/get-unity/download ML-Agents Setup (Python):...
Unity ML-Agents - Watch this first! (Is this framework for you?)
Переглядів 20 тис.3 роки тому
Before learning Unity ML-Agents, you should watch this to make sure it is the framework for you. Watch this after: ua-cam.com/video/_9aPZH6pyA8/v-deo.html And then this: ua-cam.com/video/2Js4KiDwiyU/v-deo.html Support me on Patreon: www.patreon.com/user?u=25285137 ML-Agents Discord Channel: discord.gg/wDPWsQT Keep in touch: SebastianSchuc7
Unity ML-Agents - Under the Hood (How it works)
Переглядів 7 тис.3 роки тому
Unity ML-Agents is a Machine Learning Framework that is deceptively easy to use. Let's open it up and see what's inside. The latest version of ML Agents supports Pytorch which is great! Support me on Patreon: www.patreon.com/user?u=25285137 ML-Agents Discord Channel: discord.gg/wDPWsQT Keep in touch: SebastianSchuc7
Is OpenAI still open?
Переглядів 10 тис.3 роки тому
OpenAI, now mostly known from GPT-3 / GPT-2 is in a very unique position as a company. I tried to cover this topic as objective and neutral as possible, still, everything said here is my opinion. I am a big fan of many things produced by OpenAI, like GPT-3, GPT-2, PPO. Their algorithm (PPO) is something I use all the time, so it's not like I don't have a bias. Please, let me know what you think...
Unity ML-Agents 1.0+ - Self Play explained
Переглядів 10 тис.3 роки тому
Unity ML-Agents 1.0 - Self Play explained
Why GPT-3 changes everything (and how it works)
Переглядів 35 тис.3 роки тому
Why GPT-3 changes everything (and how it works)
Unity ML-Agents 1.0 - A.I. Shooting Game (FULL WALKTHROUGH)
Переглядів 32 тис.3 роки тому
Unity ML-Agents 1.0 - A.I. Shooting Game (FULL WALKTHROUGH)
Unity ML-Agents - 5 things you didn't know about - Version 1.0+
Переглядів 9 тис.3 роки тому
Unity ML-Agents - 5 things you didn't know about - Version 1.0
ML-Agents 1.0+ | Create your own A.I. | Full Walkthrough | Unity3D
Переглядів 63 тис.4 роки тому
ML-Agents 1.0 | Create your own A.I. | Full Walkthrough | Unity3D
Unity ML-Agents 1.0 - Training your first A.I
Переглядів 112 тис.4 роки тому
Unity ML-Agents 1.0 - Training your first A.I
Building A.I. Youtube Title Generator (Machine Learning)
Переглядів 2,6 тис.4 роки тому
Building A.I. UA-cam Title Generator (Machine Learning)
How I created an A.I. that learned to WALK
Переглядів 10 тис.4 роки тому
How I created an A.I. that learned to WALK
He created the first Artificial Intelligent Robot (Mechanical Tortoise)
Переглядів 1,4 тис.4 роки тому
He created the first Artificial Intelligent Robot (Mechanical Tortoise)
5 Hacks to speed up ML-Agents Training in Unity3D
Переглядів 16 тис.4 роки тому
5 Hacks to speed up ML-Agents Training in Unity3D
The Year Artificial Intelligence changed forever
Переглядів 8 тис.4 роки тому
The Year Artificial Intelligence changed forever
DEEPMIND A.I. BEATS 99.8% OF ALL STARCRAFT 2 PLAYERS
Переглядів 2,4 тис.4 роки тому
DEEPMIND A.I. BEATS 99.8% OF ALL STARCRAFT 2 PLAYERS
The Man who forever changed Artificial Intelligence
Переглядів 24 тис.4 роки тому
The Man who forever changed Artificial Intelligence
so. is it possible to take what the ml agent learned and add it to ai? or am i stuck with the agent trying to learn?
WOW crazy man Thanks <3
The photo you show between 1:13 and 1:23 isn't Rosenblatt's "Mark I Perceptron". It's the IBM "Harvard Mark I" or "Automatic Sequence Controlled Calculator (ASCC)", a general-purpose electromechanical digital computer installed in 1944 and dismantled in 1959 (according to Wikipedia). Quite different from the Perceptron.
which Reinforcement Learning algorithm/techniques are used in this project ??
Hi Again! I am have made a hide and seek environment with a single seeker agent and a single hider agent. What I would like to do is to use the ray perception sensor 3d to give rewards depending if the hider is in the seekers view or not. What should I use for this, as the resources for the subject are rather scarce. Also, are the tags used in any way? for example if the seeker sensor hits the hider sensor(tag 2) then the seeker gains rewards and respectively the hider loses rewards
How to make a ai learn to walk
The AI when penalty is too harsh 😔💥🔫
this doesnt work any more; ie loading the example assets causes an error
can we do this multiple like 9-10 classes?
And now it's normal that I've just finished another normal night coding with GPT-4 Turbo...
I will very gladly take more content like this over all the ChatGPT junk that searching for stuff like this turns up these days
When we can expect updated tutorial? :)
What tool are you using?
What software is that?
When we upload the image of Churros, does the dimension has to be 224 x 224? Coz I just uploaded churros image of different dimension and it is giving me Samosa answer even though it is churro pic.
Thanks for the video. How long it took you to make this setup?
Finally, now I can play Minecraft with an AI because I don't have friends :3
When i add the ray perception sensor I cant see it at all in the scene view. Does anybody know how to fix this problem?
Amazing video invoking my curiosity
Can you make video highlighting gpt 4
just give him the cheese please he just wanted some cheese
Ahead of its time
on minute 2:00: Did not age that well 🤣
How do u get the model with the joints? And how do u like program the ai so it learns?
blud just wanted some cheese
Mate, is this Aussie?
I thought I left a comment on this a year ago but apparently not. This is truly amazing stuff and is very well explained.
How's this project going?
According to GPT lol this behavior was only the result of stimulus through the photo-sensor and perhaps a reverse to be triggered by the contacts meeting when it bumped into an obstacle. I was like, well how would it have "known" to steer towards the light? It said that it was designed in such a way that if light was shining towards it from the side, it would just cause one motor to receive more voltage than the other and therefore uneven torque favoring one side. GPT is so remarkable. Anyhow, is this correct? I find this really cool.
Well, I now know the info was a little misleading I had gotten from GPT, so I wanted to clear that up. They indeed did have analogue perceptrons.
@@rickylehr9284Machines, son.
Yes that sort of true they can't learn beam robotics followed this idea of thinking into more modern day, their walkers were quiet capable using motors hooked directly to a oscillation circuit. If a leg gets stuck for instance it will increase the motors resistance and slow down the oscillation giving that leg more time to get out giving it a simple walking adaptation system and some rare ones can store information in a capacitor. As for them not being AI it is hard to say what level intelligence starts and it just being stimulus don't forget the human brain is an analogue device so neural networks are making digital systems act as an anologue system so anologue stimulus approach has come back in a way.
How many steps did you take to make the agent complete the race smoothly?
dosent work as always
Man you are awesome!
You have such a pretty voice and your videos have been so helpful!
Can you do a video on how to work with the curriculum feature?
Legend
Love this series, i find it so amazing that they got the idea from looking closely at the human brain, astounding!
😀
i got 32 error saying no suitable method found to override. what can i do to fix these bugs? i need help
Great video and great project, really enjoyed going through your project, very insightful. So I need some help, I am training my agent for almost a month now (as in training stopping and trying something else), I can't seem to get to the max level, the best agent I trained simply stops me from wining, but rarely go for the win, like if he has two in a row, and you have two in a row, he blocks you instead of wining, and I get to this point on the first 700k ish steps, if I trained longer, the agent goes nuts and is no more an AI and more like I do 1 then 2 then 3 player. I would appreciate the help if you gave me some advices or anything you done under the hood with your agents. And you mentioned in the video that your AI isn't perfect, any idea on that ? can't I reach a perfect player with self-play ?
I was waiting for the quote attribution... to say that this person is really unaware of ChatGPT and… well, unaware of AI, but then it hit me: the author is ChatGPT itself, brilliantly illustrating the paradox of being both aware and unaware of its own existence. How ironic!
Take the guns
XOR : imma ruin this whole perceptron career .
1:57 unless youre in thailand.
Why is a picture of John McCarthy the thumbnail image for this video? He is not mentioned at all
HOW THE BLOODY FORK.. excuse me... DO YOU TRAIN YOUR AGENTS? My computer cant last that long. Please a video on how you do it? Can you make a video for hosting on a dedicated server with gpu instance (preferably linode)?
Missed the deadline, but eager to see what people have come up with. Community-outsourced content FTW ;)
Oh Schuchmann! ... trollo. : )
10:50 "Which is heavier, a mouse or an elephant?" *Which is heavier, 1kg of cotton or 1kg of iron ???*
Ur sad voice 😂 😂 😂
How many behavior parameters can we have, and how can we get the actions from these specific parameter components?
I was really into this and super eager to subscribe but this 12 minute video just didnt cut it for me because it really all just felt like a high level overview with basic explanations which is not what I was expecting to learn when I clicked on a video called "Training your first A.I" since the video didnt walk me through actually creating a new agent and behavior and then training the AI to do that task as would be assumed from the context of the video. When you upload a more complete tutorial I will like, subscribe, hit the bell icon and tell all my friends to do the same.
this tutorial has more info on how to do that. ua-cam.com/video/RANRz9oyzko/v-deo.html&t. I made it a month ago so everything is up to date.