Samuel Arzt
Samuel Arzt
  • Видео 8
  • Просмотров 14 533 920
AI Learns Parallel Parking - Deep Reinforcement Learning
Big thanks to Hostinger for sponsoring this video! Go to hostinger.com?REFERRALCODE=1SAMUEL08 and get 20% off your hosting plan.
I'm really glad how my new portfolio worked out! samuelarzt.com
Also check out the original parking video: ruclips.net/video/VMp6pq6_QjI/видео.html
Two AI fight for the same parking spot: ruclips.net/video/CqYKhbyHFtA/видео.html
Neural Networks explained in a Minute: ruclips.net/video/rEDzUT3ymw4/видео.html
Subscribe for more content like this:
ruclips.net/channel/UC_eerU4SleeptEbD2AA_nDw
Follow me on Twitter for more frequent updates on my projects:
SamuelArzt
Last time we trained an AI how to park (ruclips.net/video/VMp6pq6_QjI/видео.html). A lot of people ...
Просмотров: 149 579

Видео

Two AI Fight for the same Parking Spot
Просмотров 296 тыс.4 года назад
After training a car to park (watch the training here: ruclips.net/video/VMp6pq6_QjI/видео.html), I decided to put another agent into the environment and let them fight for the same parking spot. I thought the result was quite fun to watch, so here is a video of it! The AI consists of a deep Neural Network with 3 hidden layers of 128 neurons each. It is trained with the Proximal Policy Optimiza...
AI Learns to Park - Deep Reinforcement Learning
Просмотров 3 млн5 лет назад
An AI learns to park a car in a parking lot in a 3D physics simulation. The simulation was implemented using Unity's ML-Agents framework (unity3d.com/machine-learning). The AI consists of a deep Neural Network with 3 hidden layers of 128 neurons each. It is trained with the Proximal Policy Optimization (PPO) algorithm, which is a Reinforcement Learning approach. Basically, the input of the Neur...
Explained In A Minute: Neural Networks
Просмотров 466 тыс.7 лет назад
Artificial Neural Networks explained in a minute. As you might have already guessed, there are a lot of things that didn't fit into this one-minute explanation. You can read my accompanying blogpost for some more details on things I might have left out: arztsamuel.github.io/en/blogs/2018/EiaM-NeuralNetworks.html If you like these kind of videos and would like to see more technical topics explai...
Deep Learning Cars
Просмотров 11 млн7 лет назад
A small 2D simulation in which cars learn to maneuver through a course by themselves, using a neural network and evolutionary algorithms. Also check out my other project "AI Learns to Park": ruclips.net/video/VMp6pq6_QjI/видео.html Two AI fight for the same Parking Spot: ruclips.net/video/CqYKhbyHFtA/видео.html Interested in how Neural Networks work? Have a look at my one-minute-explanation: ru...
Der A*-Algorithmus - Grundlagen
Просмотров 21 тыс.8 лет назад
Dieses Video erklärt die Grundlagen des A*-Algorithmus. Die in diesem Video gezeigten Programme sind auf meiner GitHub Seite zu finden: github.com/ArztSamuel/A-Star-Pathfinding Weitere nützliche A*-bezogene Quellen: Beginner's Guide by Patrick Lester: www.policyalmanac.org/games/aStarTutorial.htm Amit's A*-Pages: theory.stanford.edu/~amitp/GameProgramming/ This video is also available in Englis...
The A*-Algorithm - Basics
Просмотров 14 тыс.8 лет назад
This video explains the basics of the A*-Algorithm. You can find the programs shown in this video on my GitHub page: github.com/ArztSamuel/A-Star-Pathfinding Other useful A*-related sources: Beginner's Guide by Patrick Lester: www.policyalmanac.org/games/aStarTutorial.htm Amit's A* Pages: theory.stanford.edu/~amitp/GameProgramming/ Es gibt auch eine deutsche Version dieses Videos: ruclips.net/v...
Blocked - Trailer
Просмотров 13 тыс.8 лет назад
In Blocked the eternal fight between humans and zombies continues. Fight for your life in this 2D-topdown shooter against hordes of zombies, replenishing your weapons after each wave. Support Blocked on Steam Greenlight: steamcommunity.com/sharedfiles/filedetails/?id=857784501

Комментарии

  • @FizzleFX
    @FizzleFX 4 дня назад

    Tesla wish it could do that.

  • @shalevforfor5550
    @shalevforfor5550 10 дней назад

    What about reverse parking

  • @ahmadjaradat3011
    @ahmadjaradat3011 13 дней назад

    Awesome video

  • @neethanvarkey
    @neethanvarkey Месяц назад

    did u use c code to program this . please this

  • @oldpop2377
    @oldpop2377 2 месяца назад

    Does that green line between the nodes mean the car has educated well?

  • @josh3.1415
    @josh3.1415 2 месяца назад

    Oh no it's more than a minute

  • @nathancampbell8064
    @nathancampbell8064 2 месяца назад

    I’m curious what causes the AI’s preference for pulling into the parking space in a forward orientation. Like, why not back in? Same reward, right?

  • @MoonGlow22
    @MoonGlow22 3 месяца назад

    İt seems they go slower when there is no other car My guess is they get rewarded for the time they are green, and so their goal That explains why they still keep racing after track is ended

  • @omarvpg4006
    @omarvpg4006 3 месяца назад

    Still better than girls parking

  • @avdhantyagi
    @avdhantyagi 3 месяца назад

    Through goes hamilton

  • @LeoKators
    @LeoKators 3 месяца назад

    now that's podracing!

  • @pikachucreation905
    @pikachucreation905 3 месяца назад

    May 2024 me video dekha hu bro

  • @hanif2285
    @hanif2285 3 месяца назад

    casual drinkers trying hard to make drink drive possible

  • @yflow5337
    @yflow5337 3 месяца назад

    That is why only one sperm out of many reaches oocytes

  • @TheRealDodgeCar
    @TheRealDodgeCar 3 месяца назад

    So what game you wanna called it Make sure its on Googleplaystore

  • @AnggaGeo
    @AnggaGeo 4 месяца назад

    So cute..

  • @AbhishekYadav-fc2we
    @AbhishekYadav-fc2we 5 месяцев назад

    Is this example of reinforcement learning?

  • @MuhammadIrtiza-ik3ng
    @MuhammadIrtiza-ik3ng 5 месяцев назад

    Practice makes an AI perfect.

  • @diagorasofmel0s
    @diagorasofmel0s 5 месяцев назад

    this is so creepy when you think about how evolution of living things also follows the same rules, survival of the fittest and mutual aid.

  • @damiangherbali6958
    @damiangherbali6958 5 месяцев назад

    In what app is this happening

  • @user-qs8dz2tr2d
    @user-qs8dz2tr2d 5 месяцев назад

    for.. humanity..!!

  • @chideraagbasiere7868
    @chideraagbasiere7868 5 месяцев назад

    This stressed me out so bad

  • @Nitishy5104E
    @Nitishy5104E 5 месяцев назад

    What the average time per attempt? 😢

  • @bezelyesevenordek
    @bezelyesevenordek 5 месяцев назад

    nice

  • @dannywhite132
    @dannywhite132 5 месяцев назад

    I feel like this isn't learning its just trial and error and remembering the inputs that did best. It's would be learning if you put it in a new layout and it still navigated the track

  • @kyung.lee.official
    @kyung.lee.official 5 месяцев назад

    Finally someone can explain NN without bullshit in 1min with a complete simple example step by step!

  • @TryHarderrr
    @TryHarderrr 6 месяцев назад

    I want to make something like this, where do I get started

    • @solaire_of_the_east
      @solaire_of_the_east 6 месяцев назад

      It is a deep learning model that requires a lot of experience in the discipline of mathematics. If it is your jam, plenty of courses are available online.

  • @arrebol-qe6zf
    @arrebol-qe6zf 6 месяцев назад

    cool

  • @alexsgarbagehut2517
    @alexsgarbagehut2517 6 месяцев назад

    Now we know the difference between a bot and a learning bot: a bot wins everytime, a learning bot needs learning

  • @theblueplanet3576
    @theblueplanet3576 6 месяцев назад

    Interesting.. It would be good to share some details like what's the input to neural net?

    • @solaire_of_the_east
      @solaire_of_the_east 6 месяцев назад

      Exactly, it would have been a lot better if they had provided details of the optimizers and activation functions they are using.

  • @lair6400
    @lair6400 6 месяцев назад

    Thank you

  • @h_j3635
    @h_j3635 6 месяцев назад

    I realized that this race was no different from our lives.

  • @mateusdsp6871
    @mateusdsp6871 7 месяцев назад

    Gooodd! :D Jesus is coming back, God bless!

  • @alex_tusinov3943
    @alex_tusinov3943 7 месяцев назад

    Те самые боты в мафии на миссии с гонкой

  • @KonvictionTV
    @KonvictionTV 7 месяцев назад

    Why does it take so many times for it to figure out? Why can't they just figure it out from the first attempt? Like what exactly are they learning? Especially when so many of them just run into the wall at the beginning. You'd think they'd all figure things out.

    • @rolandkiss4958
      @rolandkiss4958 7 месяцев назад

      It feels like just brute forcing at this moment

  • @xilechronicles
    @xilechronicles 7 месяцев назад

    Sooo boooring. Play old school mmorpg "Xile Chronicles" ❤ Find it on android stores😊

  • @shalevforfor5550
    @shalevforfor5550 7 месяцев назад

    This is how viruses are work right

  • @csuulok
    @csuulok 7 месяцев назад

    Only 15 generations?? Thats insane!

  • @trandautomation
    @trandautomation 7 месяцев назад

    Deep Learning? Not RL?

  • @Technaton_English
    @Technaton_English 7 месяцев назад

    Which GPU?

  • @user-eq7rw4sh1d
    @user-eq7rw4sh1d 7 месяцев назад

    Whoa!That' amazing!

  • @shivamdubey4783
    @shivamdubey4783 7 месяцев назад

    Have you used genetic algorithm also

  • @buatt
    @buatt 8 месяцев назад

    This is how currently Tesla cars are learning to drive.

  • @nooranshdhaka
    @nooranshdhaka 8 месяцев назад

    We need you back.

  • @plane_a
    @plane_a 8 месяцев назад

    This is how gpt useful now)

  • @Nanashi_Munchlax
    @Nanashi_Munchlax 8 месяцев назад

    精子だった頃を思い出す

  • @SunAndMirror
    @SunAndMirror 8 месяцев назад

    This is footage of my mothers fallopian tube the day i was concieved

  • @fellpower
    @fellpower 8 месяцев назад

    A computer that can calc billions of possibilities in 1 sec cant do it, after 10k tries. but i have to do it in 20hrs of driving school....

  • @davidkorinek8666
    @davidkorinek8666 8 месяцев назад

    I think the punishment is not high enough. 😂

  • @mubarakdalvi9004
    @mubarakdalvi9004 8 месяцев назад

    i am currently learning data science and i want to know does deep learning depend on data frame or any other sources, means data can be in form of code so you train your code to be perfect so you create file of code so you can train your code and then predict your out or what i dnt get it, cause i haven't started deep learning yet but when ur i watched ur video i said that how dl work and this is awesome