AI beats multiple World Records in Trackmania

Рет қаралды 2,957,028

4 ай бұрын

I trained an AI in Trackmania with reinforcement learning, and made it compete against human World Records on 3 different pipe tracks.
Between research, programming and editing, these videos take a long time to produce. Any support on Patreon will help me to spend more time on that in the future :)
• Patreon : / yoshtm
Contact
• Discord: yosh_tm
• Twitter: yoshtm1
• Mail: yoshtm.yt@gmail.com
The maps shown in this video can be downloaded on TMX and played in Trackmania Nations Forever:
• 1) One Hella Long Pipe (It requires TMUnlimiter !) - tmnf.exchange/trackshow/8484272
• 2) Calm Down - tmnf.exchange/trackshow/1293088
• 3) Are You Serious ?! - tmnf.exchange/trackshow/5152869
Wirtual made a nice video about his world record on the track Calm Down, don’t hesitate to watch it : • How I Beat Trackmania'...
More generally, if you want to learn more about this game, check out his streams and videos, he makes fantastic content !
• twitch.tv/wirtual
• / @wirtual
You can find a list of the musics I used at the end of the video. Special thanks to Beik Poel for allowing me to use their song En aften ved svanefossen : • En aften ved svanefossen
Thanks to Donadigo, for TMInterface !
donadigo.com/tminterface/
All Trackmania Nations Forever tricks TIERLIST - Fliks.
• All Trackmania Nations...
Into The Breach vs. Karmine Corp | Semifinal 2 | World Championship 2023 - Trackmania World Tour
• Into The Breach vs. Ka...

Пікірлер: 5 000

@yoshtm 4 ай бұрын

Thanks for watching! Some additional details not explained in the video, which might help to better understand the irregularities observed: - I didn't show all the AI inputs in the video to keep things simple. In reality it has access to more information, such as x,y,z velocity, velocity rates, roll-pitch-yaw rates, etc. But maybe it's still missing some crucial information, it's hard to know. - The irregularities observed are not due to hardware or framerate issues. All the things I'm showing in the video are made with a tool called TMinterface. This tool allows me to inject action commands into the game at precise timings, in a way that is 100% repeatable. The same sequence of action on the same map will aways lead to the same outcome, even on a different computer. It's completely deterministic. And it's used a lot for Tool Assisted Speedrun (TAS), you can find many examples of that on youtube. I have a bit of extra footage that I couldn't fit into this long video. I plan to post some of this extra content on my Patreon in the next weeks. Any support there is a great motivation to keep making these videos :) patreon.com/Yoshtm

@andrewbrown2038 4 ай бұрын

Do you know if the inputs are completely deterministic independent to time? If you run the same set of inputs offset in time, will the result be identical?

@yoshtm 4 ай бұрын

@@andrewbrown2038 If you are speaking of ingame time, it has an effect. Starting to accelerate at t=0.00s vs starting to accelerate at t=0.01s are two different inputs, which will have different outcomes, as shown in the video around 32:00. But it's still deterministic: starting to accelerate at t=0.01s will always have the same outcome in successive races.

@gofai274 4 ай бұрын

Where to watch AI playing games i am sooooooooo boooooooooorrrrrrrrrreeeeeeeeeeeddddddddddd!!!!!!!! It plays in real time sometimes... Damn wish there was good AI to play with already :(

@mattanderson5239 4 ай бұрын

So from what I remember about neural networks, each neuron of a neural network takes the inputs as numbers multiplied by weights and passes them through a function to output another number to be taken as input once again. Now I don't know the specifics of your neural network such as what activation function you are using, but combining this with how your outputs of the network will affect the inputs after any amount of time, changing a specific input at any specific time, even slightly, would create a chain reaction that would get you your random variations after any period (usually 1 or 2 seconds) as seen at 18:44. Thus, the random variation is not necessarily a problem of incorrect inputs or anything but rather an aspect of neural networks. Perhaps I am misunderstanding some way, or maybe you had already answered this question, but I believe that is the cause of the random variation.

@yoshtm 4 ай бұрын

@@mattanderson5239 a few seconds later, around 18:50, I show that even when I force all cars to use the exact same sequence of actions after the perturbation, the randomness is still there. Maybe the neural network is part of the problem, but according to this result, it can't be the only source of randomness

@tylerlarsen1842 4 ай бұрын

I came here to see AI destroy humans on Trackmania, but I got a 37 minute essay on how pipes in Trackmania are an exercise in chaos theory instead. 10/10

@xenmaifirebringer552 4 ай бұрын

While still seeing an AI smash the human WRs! I see it as a win-win scenario 😊

@wroomwroomboy123 4 ай бұрын

Dude took the most niche and least enjoyed playstyle ... to make a 10/10 documentary of his project. Easily best TM youtuber atm

@rank944 4 ай бұрын

thanks for the save

@MaxwellCatAlphonk 4 ай бұрын

This is better than school 🤣

@Ice190236 4 ай бұрын

Not only Pipes are an exercise, obviously the... (what was it, eo3?) track as well. 😂

@night0x453 4 ай бұрын

PhD Student in Deep RL here. The behavior at the end seems mathematically chaotic and might lead you to the conclusion that you cannot predict deterministically the behavior of the car. However, that does not mean you can't improve the performance by a long shot with a few simple tricks. You are doing model-free reinforcement learning, which basically means that you don't need to predict what is going to happen exactly (and extremely hard here), you just need to figure out what actions are the best in this situation. In most environments, this is actually much easier to do (and a reason why learning an accurate model is often harder than just straight up optimizing for performance with model-free). Second problem is that you are using an RL algorithm and RL assumes you are in a Markov Devision Process, ie that you have full observability of the world, ie that the probability of taking the next action ONLY depends on the current information that you have, and not the past, ie that adding information about the past does NOT help ou make better decisions. HOWEVER, you are NOT in an MDP (Markov Devision Process) here since you lack critical information about the state to take decision, as many pointed out. In practice, what you can do will most certainly improve performance: - either manually add all the missing info (speed, rotations, ...) - or more simply, create a concatenated vector of the previous N past states and inputs and use that as input to your NN (LSTM/RNN works too but is more complex and often not necessary when we are dealing with a few steps of history). - Use more layers (deeper network = more complex function can be learned by the network) - Sticky/random actions: introduce manually randomness in the action of the agent. For example repeat randomly the past action with a small probability. Do NOT decrease this probability with time. What this means is that the agent will have to learn the whole time in a now stochastic (ie random) environment and will have no choice but to cope with the inherent randomness that you added, making it largely more robust than an deterministic agent. This is a common flaw of fully deterministic agent in these kind of environment btw. - Look at extensions like Max Entropy RL (Soft Actor Critic for instance), where you both maximize reward and entropy of the policy. Some papers proved that it makes the learned policy more robust to out-Of-Distribution perturbations, ie perturbations that were never seen during training. In your case, this will help the agent lean "recovering behaviors" to recover from deviations caused by physics bugs/randomness/whatever. - Max Entropy RL will also help with exploration a lot, which might help the AI be more creative in finding solutions. - Look at other tricks that improve performance "for free". I don't know how much you implemented already, but a basic start is looking at Rainbow DQN, PPO.... But you are already using it probably..?

@raskreia8326 4 ай бұрын

I am a PhD student myself but in Sexology and I have nothing to offer here.

@7dedlysins193 4 ай бұрын

@night0x453 Hi do you have any course recommendations/ or youtube channel for beginners in ML/RL?

@night0x453 4 ай бұрын

@@7dedlysins193 look at Sergey Levine course on Deep RL, it's one of the best out there to get started in deep RL/robot learning

@beingatliberty 4 ай бұрын

lol@@raskreia8326

@FredDufresne 4 ай бұрын

I'm not a PHD student in either sexology or Deep RL, but I can tell you that this wall of advice right here is sexy af.

@Arturius1987 3 ай бұрын

That ending sequence was an absolute joy. Prefect syncing with the music

@miglol9279 8 күн бұрын

witrual should trademark that line LOL

@dobryden7196 Ай бұрын

Gotta say man, your visualization skills are insanely good. Great job. Perfect execution.

@MagicWazam 4 ай бұрын

28:40 was a comedic masterpiece, the editing, the pacing, incroyable!

@evenleven9205 4 ай бұрын

Absolutely poetic.

@hairyrosette9642 4 ай бұрын

God, I wanted to see turtle on pipes so badly 😢

@96thelycan 4 ай бұрын

"Ive Tried though" how many hours we're lost to these 3 words

@camillaericajuliarandal332 4 ай бұрын

😂😂 I was laughing so hard at that point

@ianweckhorst3200 4 ай бұрын

Genuine comment though, if you give the ai full knowledge of the decimal precision you’re getting, it might actually be able to do something about it, give it the smallest of minute details, only then can it do something about it

@sreynoldshaertle 4 ай бұрын

Roboticist here. Some regions of configuration space are more chaotic than others; the demo you did where you perturbed a car that was about to fall off demonstrated a smooth increase in success rate as you moved the perturbation further away from the point at which outcomes diverged, suggesting that the failure was in a fairly narrow chaotic _region_ and moving out of that region made the system more stable and enabled more consistent success. This is what unnamed did to complete the jumps track: Found a strategy that consistently avoided highly-divergent regions of configuration space. One problem is that, if the display of the inputs is representative, the AI doesn't have the right sensors. Specifically, the AI has its position and speed but not its _velocity_; it doesn't know how much of its speedo reading is in each component of down the track vs. across the track vs. away from the track. More critically, it also doesn't have roll-pitch-yaw rates. It only knows its current orientation, not the rate at which its orientation is changing. I suspect that the car also needs to know the difference between a simple pipe elbow, a tee, and a cross. Even if the extra geometry isn't in the direction of its travel it'll almost certainly still interact. Finally, consider increasing the numerical precision of your network. This is one of those cases where running floats vs. doubles might actually matter.

@granite_planet 4 ай бұрын

Adding rate inputs, or more generally some memory of past frames (recurrent neural networks), should indeed improve the results. Unless Yosh is already using those and has just kept the narration simple. :)

@solsystem1342 4 ай бұрын

I hadn't noticed that but that's a great point. If you think of the ai as trying to force configuration space into the few fast values where it doesn't fall off being able to actually see the full space can be a great help. Otherwise it becomes impossible to figure out the optimal action. Of course there might be performance and training concerns if you take it to the extreme but experimenting with adding more inputs could be a good way to see what improves consistancy.

@Bruellhusten123 4 ай бұрын

fully agree - the AI needs more information to make the right decision. And I don't think giving it all the information would solve all the problems. Because the game engine itself is just a simulation of real world physics. There also is a tick rate and some approximations/simplifications being done. Some things will be hardcoded like material coefficients for energy conservation during different types of bounces (side of the wheel against a wall is vastly different than bottom of the wheel to road/pipe surface you land on). Imo yosh needs a well trained model on the game engine itself. In a way that the AI is able to predict the result of each action it makes.

@asterpw 4 ай бұрын

I think this is it. The inputs look like they are suited for a track with a 2D layout like in his last videos but these pipes need a better 3D understanding of angular momentum. The AI is doing the best it can with a blind spot in its vision.

@CouchPotator 4 ай бұрын

In previous videos, Yosh explained he uses the last X frames in addition to the current frame as input, so I'd assume he's doing the same here. (At least I think Yosh does, it's hard to remember which AI video does what :E)

@kirdow 4 ай бұрын

I’m not a TrackMania player but I love watching TrackMania videos. I also love game AI vids, and this video really did it for me. Being fascinated by chaos theory I’ve been screaming at my screen throughout the whole video ”bro it’s chaos theory”, and then you even mention it yourself. Top tier content, it’s what I’m here for. You just gained a new subscriber 😄

@saturnhex9855 3 ай бұрын

Haha same! I'm into chaos theory too, so the second he was like "it's deterministic, but it still feels so random...little differences in input..." I was yelling at my computer!

@tomepedro2560 4 ай бұрын

This is one of the best videos I've ever seen. Amazing work!

@xardenas2636 4 ай бұрын

28:50 “Now let’s have the AI drive upside down” had me rolling. For a hot second I really thought you were about to blow my mind once again 🤣

@dabestgrimmreaper4 4 ай бұрын

Bro, i was like no fkn way he beats wirtual again upside down and then does that backwards just to show off that wirtual is just driving like a grandma on that track lol

@ArmageddonNerd 4 ай бұрын

My stunned disbelief in that moment was at peak levels. I 100% thought it was going to happen haha

@xardenas2636 4 ай бұрын

@@ArmageddonNerd I’m glad it wasn’t just me 🤣

@haqtanefe 4 ай бұрын

LOLL

@BLENDITE 4 ай бұрын

HE EVEN HAD TRAINING PREPPED FOR IT 🤣

@wurrbm 4 ай бұрын

22:48 "THE AI GOT THIS RUN" omg even the song, that's great man

@theaypisamfpv 4 ай бұрын

but then, hefes... heu i mean the AI got this run...

@wbrim 4 ай бұрын

@@theaypisamfpv but then AIHefest got this run:

@sahebiyasin6988 4 ай бұрын

"And then hefest got this run"

@sca04245 4 ай бұрын

Been looking for this comment

@GameristicForce 3 ай бұрын

666 LIKES?????????? Every other comment has like 15!!

@Hurricane6220 4 ай бұрын

What an insanely good video! I haven't even played Trackmania in many years... but the way you explain the details and visualize everything just makes this soooo interesting! 😌👏👏👏

@maxouilletm 2 ай бұрын

I saw you on the free zone speed server last week

@maxouilletm 2 ай бұрын

I saw you on the server freezone speed last week

@gui6987 3 ай бұрын

Incredible video, one of the best I’ve seen in a long time, thank you !

@yuxufgVOD 4 ай бұрын

The song that played when you showed Wirtual's WR was just peak video making lol

@bequerhernandez8487 4 ай бұрын

Haha I was looking for this comment 😂

@jamport5973 4 ай бұрын

Time stamp?

@1BenPro 4 ай бұрын

@@jamport59739:00

@skydemon3423 4 ай бұрын

9:03? you probably already seen it@@jamport5973

@TheChaos1123581321 4 ай бұрын

Then the AI got this run,

@happycryingcat3101 4 ай бұрын

Yosh: "I am going to lock your breaks and acceleration so you cant slow down" AI: "Pfft check this out"

@binz2056 4 ай бұрын

and that it still figured out how to manage its speed with airtime despite those limitations. and he says the AI isn't creative.

@bertram-raven 4 ай бұрын

@@binz2056 That is entirely deterministic. The increased airtime led to success. It is just maths masquerading as creativity.

@binz2056 4 ай бұрын

@@bertram-raven call it whatever you want. But it is an interesting workaround given the constraints.

@imoud47 4 ай бұрын

Yosh: "I am going to make you drive backwards then" AI: "Pfft check this out"

@thedave1771 4 ай бұрын

@@bertram-ravenisn’t that true of all creativity that generates something that can be represented mathematically?

@AlmightyRawks 4 ай бұрын

Apart from the excellent research, I'd just like to comment on how amazing the editing was for this! That kept it interesting AND funny!

@cristianguatuzmal3149 Ай бұрын

this video is about something very technical, yet, it touches something deep inside me, something more emotional rather than rational. I sincerely congratulate you on this, I feel this turned out to be more philosophic and even artistic than I expected it to be. (also, huge detail to use wirtual theme and editing style, it adds a whole new emotional layer)

@qweytr9964 4 ай бұрын

Your analysis about chaos theory is exactly what is going on. And TrackMania's chaotic physics are very familiar to me from another TrackMania project. The best way to control that chaos is by going slower, as the deviations caused by the bugs are less severe at lower speeds. Perhaps giving less reward for speed and more reward by having a high chance of getting far would help with consistency.

@BoxTM 4 ай бұрын

In fact, that's probably why the backwards driving cars were much more consistent. There's not as much deviation in speed when they're going backwards, so any bug has less impact and the performance reward is more focused on getting far too.

@HECKATE 4 ай бұрын

I dealt with similar issues in some TAS work with Neon White - in that case, small variations in frametimes would lead to different interpolation of player movement (even with fixed tickrate physics, they still interpolated the actual player position based on frametimes), which would add up to significant positional desyncs occasionally. Vsync helped reduce the frequency of this (as it would better sync frametimes to monitor refreshes), but it still wouldn't fix it entirely. sometimes, it's just about the little things.

@iCrimzon 4 ай бұрын

I dont wanna say one thing is the reason for the other without 100% confidence but going forward = more speed = more chaos, backwards = less speed = less chaos, coincidence?

@Bruno-cb5gk 4 ай бұрын

@@BoxTMI'm only familiar with the basics of chaos theory, but I feel that trying to find the Lyapunov time of the system could be a good first step.

@kristofp2383 4 ай бұрын

@@Bruno-cb5gk I think you will find such a calculation not only to be impractical, but next to impossible to perform… The Lyapunov time will only be a few frames/instances of the physics engine refreshing. Anyway, we know all the hidden variables here as he showed in the video, so there isn’t much point since it still is deterministic and we can actually view all the precise starting information right, why can’t we just let the ai view that data as well? Correct me if I am mistaken please.

@bettaTM 4 ай бұрын

So you’re saying my world record fails are actually the butterfly effect’s fault…

@gabrielandradeferraz386 4 ай бұрын

most certainly. it also means that all of thousands of times you failed were simply bad luck, and everyone to ever succeed at a world record simply got lucky, and this is not at all a skill issue

@shadowlord0162 4 ай бұрын

funny how there is some truth to the sarcasm@@gabrielandradeferraz386

@Vaaaaadim 4 ай бұрын

New coping mechanism just dropped

@winumoritribe8425 4 ай бұрын

@@Vaaaaadim the most mecchy player. coping 100

@_Ekaros 4 ай бұрын

Or your successes...

@luigibeccali2840 2 ай бұрын

Thise video (and all of the other videos of yours) are a true masterpiece.

@enderdavis591 Ай бұрын

Hahaha the In The Hall Of The Mountain King montage at the end was really a master stroke of editing. Well done, as always.

@TheWizardsOfOz 4 ай бұрын

Waiting for all the Trackmania KZfaqrs to react to this.

@shrimpscampi9320 4 ай бұрын

sex update when?

@Guy.1. 4 ай бұрын

Yeah

@Guy.1. 4 ай бұрын

Yeah

@peskybirb7217 4 ай бұрын

Yeah

@Cyroxbtw 4 ай бұрын

Yeah

@BernhardMillauer 4 ай бұрын

your visualizations are just amazing, thank you for all the editing!

@itmsolver 4 ай бұрын

Amazing video! Interesting work

@Controllerhead 4 ай бұрын

Yeah, these 1000+ AI car attempt renders are just visual candy.

@firepaint33 2 ай бұрын

What an extremely high production value we have here. The edit, the music, the time and dedication oooozzzzee excellence in the art. Well done to all involved. This is a patrion worthy production for sure! Keep it up and hope to see more at this level. And yes it strengthens understandings in both chaos theory, and agnosticism lol. Mahalo

@AdHocF1 Ай бұрын

Super travail, mec, je pense qu’on attend tous la suite avec impatience. Et à mon avis, tu n’auras pas d’autre choix que de te lancer dans la théorie du chaos :p

@NaThingSerious 4 ай бұрын

Wow, this was not only informative and entertaining but a cinematic masterpiece with perfect pacing and music selection and everything, I didn’t even realise it was 40 minutes long until after I finished watching it!

@mooritzzLP 4 ай бұрын

thats exactly what i thought too!

@lobban2 4 ай бұрын

It was 40 minutes?!🫣

@erikgaynard5607 4 ай бұрын

The butterfly transistion at 17:27 is a pure masterpiece! The many small things u do while editing dosen't get enough appreceation! Congrats on a beautiful video!

@Nexovus 4 ай бұрын

Beat me to it.

@rolandgeb9970 4 ай бұрын

Me too, wanted to say the exact same thing 😊

@mr.raphael1507 4 ай бұрын

“Foreshadowing is a literary device in which a writer gives an advance hint of what is to come later in the story”

@paillat 4 ай бұрын

I was so happy when I noticed it

@katzo1558 4 ай бұрын

What if it's linked to the Butterfly Effect in chaos theory? Nice detail!

@kennethpetronis2464 3 ай бұрын

I’ve never played track mania. Only watched a handful of these videos and all I have to say is WOW! this video was amazing! Please keep making them.

@Forgivz 3 ай бұрын

Incroyable ta video, ca ta pris beaucoup de temps comme tu la dit bravo pour ton travail de fou

@dinoscheidt 4 ай бұрын

16:32 and there he is, introducing brownian motion in a virtual world to isolate entropy. Good job!

@kristofp2383 4 ай бұрын

Mathematician here, I fully agree with your chaos theoretic conclusions, and I think the pipe experiment at the end strongly supports this (as you realised). Why not give the ai control after a random amount of time has passed at the beginning? I think it is really interesting, I would like to see if the ai could somehow abuse the chaos theoretic nature of the situation - I remember a video about a Pokémon ai that figured out some elementary RNG manipulation so that it wouldn’t get grass encounters :D If you want to talk actual mathematics, then I think perhaps the KZfaq comment section is inadequate - I cant answer every question either, I only took one chaos theory class as a grad student, but perhaps we can try to understand some things.

@Life0 4 ай бұрын

I'm a mathematician specializing in Chaos theory, AMA

@axerity9212 4 ай бұрын

@@Life0 does chaos theory apply to women because i never seem to understand what the hell is going on

@phimuskapsi 4 ай бұрын

What I'm curious about as a programmer, thinking about this issue, is if for whatever reason the physics system is tied to a clock/tick rate that could potentially cause that bizarre result where a certain delay always means success. Like if you tried the sim on another machine would the delay carry over? Is it a processing limitation? There could also just be mathematical issues within the physics engine itself; e.g. If you hit the ground hard, and lets say it spikes the suspension values to some crazy number, maybe the engine tosses values over a certain threshold or under another. @Yosh I wonder if you could teach it what stability "means", and set a reward system up for "good angles" that it finds for drops that are more likely to produce consistent results.

@gabrielv.4358 4 ай бұрын

This game is insane. I could never manage to complete it better than yellow maps

@VocalMabiMaple 4 ай бұрын

@@Life0 what are your favorite examples of everyday phenomena that involves chaos theory?

@Martin3559 4 ай бұрын

Man, you have the best vidios on AI reinforcement learning on the whole internet, kudos!

@user-xq8np2to3y 3 ай бұрын

Incredible video. You have gained a fan. And I love maths and AI. So can't wait to start studying it

@EdgeTM 4 ай бұрын

Thanks for using my map :D I have heard people saying that Trackmania is a deterministic game hundreds of times. I think from now on I will prefer to use the word "chaotic" :D

@yoshtm 4 ай бұрын

Heeey cool to see you here, I had a lot of fun in the replay editor with this map :D And gg for finishing such a map, it must be painful ahah

@yinyang2385 3 ай бұрын

I want to see a merger with the world of "micromouse"

@no-no-noku 4 ай бұрын

The editing, the music choice, the compactness of the information while still remaining coherent, the pacing. This video is a masterpiece! Even small things like using the butterfly-shaped pinhole transition when bringing up that topic for the first-time, or how the hue of the cars match up with what generation they're from, so many little details that I could write up my own essay detailing every little thing you put extra effort into. I was enraptured the whole time. Bravo!

@J3TL4GG 4 ай бұрын

i dont even play this game, this video is a masterpiece!

@Xsiondu 3 ай бұрын

Your comment perfectly captures the things I noticed most. The continuous audio and visual cues where important concepts were being introduced and then the release of focus where it was ok to do so. This is just as interesting as the stuff I remained engaged enough to learn due to the editing and composition. A god damn MASTERCLASS!!!!

@dazeoliver 3 ай бұрын

@@J3TL4GG neither do i yet i sat herre and watched the entire thing because it was just that entertaining!

@J3TL4GG 3 ай бұрын

@@dazeoliver this that perfect 3AM video to watch

@series161 Ай бұрын

i was going to write this post. thank you for doing it for me ^^

@andreasjurack9068 Ай бұрын

Wow. Man, this video is awesome. Not only the content, but especially how its presented. Im impressed, to say the least

@takeone22 12 күн бұрын

Among all the AI videos I have watched, this one was by far the most interesting. Your storytelling was spot on. Thank you so much! I really enjoyed it.

@itsmethep_ 4 ай бұрын

Unironically one of the best track mania / AI beats things video I have ever seen, the visual representation of everything and your clear explanations made this so satisfying to watch. Like the showcase you did when explaining the randomness of tiny input changes was so well done and it is really clear how much time you put into this topic and video. Also your humor ist just amazing lmao

@George_vv 4 ай бұрын

I love how the AI on the 65km pipe track looks like it's just blissfully skipping along while the human records look like they're terrified. it's so silly.

@stephendubarry6260 3 ай бұрын

Possibly the best video I have ever watched on KZfaq. Well done!

@ickystickynicky4620 4 ай бұрын

I like how yosh's attitude towards the learning model went from "We shouldn't have to sacrifice consistency for speed" to "Fuck it we ball"

@CorzoTM 4 ай бұрын

yooo such an honor that you used my map for the 3rd part ♥really enjoying your AI videos :) keep up the great work!

@yoshtm 4 ай бұрын

Ooooh hey corzo nice to see you here!!! I remember playing with you on the tm2 rpg titlepack sever many years ago :) I'm glad you liked it!!

@series161 Ай бұрын

amazing. thank you for the effort! incredibly well made video!

@timvanlanen8868 3 ай бұрын

Amazing video! I love the editing

@goffe2282 4 ай бұрын

This is really stellar work. Not just the AIs but the videos. I'd equate these with the likes of Summoning Salt who are just very good at not only doing stellar work themselves (speed runs in his case) but also present it in an enjoyable and captivating fashion. I have not binged watched your material, but from the videos I have seen this is some of your best work. Keep em coming.

@yoshtm 4 ай бұрын

Thanks a lot

@hakaki7280 16 күн бұрын

I have learned and remembered more by this video than for last ikd how long, but I really think that what you do is amazing, take care.

@asmkwo 4 ай бұрын

This is the pure essence of youtube, why this platform is amazing. This video is an absolute master piece, I am forever impressed, congrats

@Zenith-ly7pr 4 ай бұрын

Your humor with "the song" and "the AI got this run" is just too damn much man. On top of being a cinematic masterpiece, it's also funny as hell. Insane video, genuinely hope you blow up beyond belief.

@ariaden 4 ай бұрын

Even deterministically chaotic systems can be controlled, in practice it is when Lyapunov exponent is small and the agent has quick enough reaction time (and steering capability) to keep control. But the touch at 34:06 looks similar to a landing bug. A sudden deviation from usual behavior. What you can do is to add another (part of the overall) deep neural network to act as a predictor. That would detect such "buggy surprises", and you can focus the score on surviving such surprises. As to making the competition with humans more fair, what you could do is to make AI commit to its actions some time (say 0.3 seconds) into future. This would lower its control, so it would have to drive slower in order not to lose control. (Edited typos.)

@Photo650D 4 ай бұрын

@Schmogel92 4 ай бұрын

@Panimioul 4 ай бұрын

👍

@Bruno-cb5gk 4 ай бұрын

Splitting the AI up into multiple smaller ones seems like the way to go. For example one for prediction, one for recovering from mistakes/bugs/chaos, one for just trying to go fast (pretty much what the AI in this video was) then a controller AI that decides when to switch between them. This approach would also make harder tracks more possible, if there would be a separate AI just for speedslides, bug slides and all the other techniques, a few prediction AIs and then a decision maker AI.

@trumpet43samurai 4 ай бұрын

But landing bugs are deterministic. They are called bugs but if you execute the exact same inputs the landing bug will happen. That’s how trackmania saves replays in fact. I think what is nondeterministic is how long it takes the AI to compute an update, so it gets a slightly different view of the world at the next step.

@FDGQQW 4 ай бұрын

one of the most entertaining videos ive seen. good job!

@joe-fc6be 4 ай бұрын

This video and others are so well-made, Yosh! Bro WTH

@matejvlcek6151 4 ай бұрын

I loved how it just randomly turned to a Wirtual documentary for a while at 21:55, even the Norwegian flag is here.

@PotatoRed148 4 ай бұрын

"In the Hall of the Mountain King" playing while all the cars rush to their demise like a cascade, is so fitting. Your videos are so interesting and it was worth every second spent.

@McNificoTube 2 ай бұрын

The video narration and edition is awesome! Keep that excellent work!! What about a video explaining how does the AI works, with some working examples, and so? I known nothing and something about AI, and I felt that you can explain all the details flawlessly.

@nullandvoid7881 4 ай бұрын

Thank you for including the music track list at the end, so many bangers in this!!!!!

@nullandvoid7881 4 ай бұрын

also just found out that Timothy Infinite released more new music TODAY 🥳🩵

@Evelaraevia 4 ай бұрын

I didn't expect there to be micro fluctuations in a still car, but its equally fascinating as it is frustrating to learn the game is so sensitive that milliseconds can, in niche scenarios, determine your success. It may not be truly random, but it's close enough that you can't account for it.

@0752756949 4 ай бұрын

But that's actually what happens to a car IRL. A straight line is a mathematical convention with absolutely no correspondence to anything in the observable universe

@youshallneverfindit8415 4 ай бұрын

OMG i was litreally watching the previous ai video and this JUST DROPPED ive been BLESSED. Actually huge timing

@dmsejin 4 ай бұрын

same!

@pufferfishjiminimi 4 ай бұрын

Same!

@_kirry_1148 Ай бұрын

What a video great job, can‘t wait to watch the next one 😅

@fudeldideldu 3 ай бұрын

Amazing how complex the simulation in Trackmania is... Never knowed. And your work ist mind boggling to me... Purely wonderful work.

@thevalarauka101 4 ай бұрын

I have no idea what's going on but it was really satisfying watching huge cascades of race cars falling off pipes, thank you!

@trucid2 4 ай бұрын

As a former kid, I concur. Watching cars fall off pipes is satisfying.

@juliensauter 4 ай бұрын

Wow. The way you paint the picture between about 17:00 and 20:00 is so captivating, wonderfully done!

@r2_bee2 2 ай бұрын

When I saw that butterfly, I thought you're gonna get deep into chaos theory. Thank you for the great video and for not complicating that topic further this video. I would love to see a project discovering the chaotic behaviour of collision with pipes, as an example, in Trackmania.

@Solapunk Ай бұрын

amazing video Yosh! Would love to learn more on how you build the AI and what you use to do it.

@maxpoppe 4 ай бұрын

This definitely is your best video ever! I've enjoyed the previous ones a lot, but this is a masterpiece! Keep going at it; the humor, the theorycrafting, your thought processes... it's all just one giant jigsaw that fits together perfectly.

@yoshtm 4 ай бұрын

Thanks a lot

@andrewbrown2038 4 ай бұрын

Software Developer with focus on ML here. Just from 32:03 you can see a lot of variance in the game engine physics calculations. This is most likely due to floating point precision error over time. To a human player, this variance isn't preceptable. Eg. one hundred thousandth of a velocity change looks the same in reality, but the different input may cause the AI to make a different decision. The AI may be over fitting into the region of noise in the signal. To make the AI more consistent and human like, I think you need to actually DECREASE the precision of these values fed into the model. That should help reduce the chaos being injected into the system. This is a balance because reducing the precision too far is bad too. A simple trick to try is to convert inputs to int8 or int16. That will give lower precision input and perform faster with aimpler math calculations. Or you can just truncate the float to the first 4-5 significant digits.

@LeonardoTaglialegne 4 ай бұрын

Great video, keep up the good work :)

@shailchoksi889 4 ай бұрын

Great video! Love the little homage to Wirtual with the sound track and "and then the AI got this run".

@RacetasClub 4 ай бұрын

What a wondeful video, the quality, the effort, the scripting, even the joke about driving turtle, totally got me, piece of art!

@elidanielsimonrada1577 4 ай бұрын

Great video. Ive totally watched it after 44sec release

@finbarshields Ай бұрын

This is such a ridiculously well-made, researched and edited video

@samygauquelin 2 ай бұрын

RL in trackmania make it very visually interesting. Thanks for creating video of this quality followed by all your thoughts over the projects.

@btf_flotsam478 4 ай бұрын

25:55 was the funniest part, no questions asked. Long but great set-up to a satisfying slapstick punchline.

@lennibastert7445 4 ай бұрын

Goosebumps indeed but I gotta the say the pun at the end that more patreon subscribers would be a great "reward" for him made me crack up even more 😂 suddenly the guy is just an AI himself 😅

@phkxv 4 ай бұрын

33:16 BUTTER FLY JUMPSCAR AWOO[

@niklasullrich6762 3 ай бұрын

Amazing channel, great content!

@skullbasher0088 4 ай бұрын

Amazing work. This was a great watch.

@MinecraftSubi 4 ай бұрын

I love your video style and how you're displaying the learning process. Thanks for the great video!

@coredoggo9686 4 ай бұрын

27:45 not only did he teach an ai to beat a wr backwards but he also was the first person to traumatize an ai

@IlGalvaa Ай бұрын

beautiful and inspirating video man

@ThePlaylistGuy-TPG 4 ай бұрын

La vidéo est très clean, le taff est fou, c'est intéressant, fun, chapeau, c'ets du très beau taff

@ChaosCreations0 4 ай бұрын

Every time I see these videos appear, I can't help but watch. It's super interesting to see the steps required to train the model and the steps you take as you iterate. On top of that, the way you've visualised all of the attempts and with the use of colour to demonstrate how close the AI is to ideal for the scenario is just pure eye-candy. I love it!

@nyghl 4 ай бұрын

This was such a masterpiece video! I’ve been following you for a long time but it is incredible on how far you have come. I enjoyed every minute of this video. Thanks. ❤

@carrottopevans 2 ай бұрын

This one got me hooked on your channel

@aaron159r2 4 ай бұрын

The music editing at the end was excellent!

@NorthTM 4 ай бұрын

babe wake up, new trackmania ai video

@sHootR450 4 ай бұрын

... just dropped

@hitchikerspie 4 ай бұрын

Comp Sci Degree here: You talk about the physics being deterministic, but at faster car speeds the collision detection from the physics engine is likely not refreshing at a fast enough rate to know the pixel/sub pixel of a collision perfectly, and is doing some level of interpolation on its calculation. This interpolation alongside your initial perturbation means the system stays chaotic, and thus explains your wildly divergent results even from the same starting position.

@hitchikerspie 4 ай бұрын

Hahaha, I just got to 32:00 and feel very vindicated on my initial assessment

@dreamliner8745 4 ай бұрын

Yeah, this could be it, or it could also be floating point arithmetic maybe, the minimal changes in values with a lot of zeros seem a lot like floating point errors

@Piokoxer 4 ай бұрын

@@dreamliner8745that's just how physics engines work. when there's gravity and something is resting, usualyl the gravity still moves it down the tininiest bit (it only has 1 tick to accelerate), then it does collision checks and snaps it back up, which obviously introduces the tiniest bit of change

@Andi89lectro 4 ай бұрын

Can you try locking the max speed of the car (AI) on different speeds or test different computers if the randomness changes?

@Erksah02 4 ай бұрын

Not only the game itself but also the hardware can cause a different result.

@bbrrrrr6553 3 ай бұрын

Intéressant, intriguant, étonnant, fascinant, et même émouvant... Quel travail d'immense qualité, merci de tout cœur !

@coolzeroh 3 ай бұрын

Best video seen this year. i really enjoyed it thx

@mimisibai 4 ай бұрын

A race without 🥕 = chaos Congrat's Yosh ! Love watching this little car grow 🧡

@juliensauter 4 ай бұрын

You make the bestvideos on Reinforcement Learning! They inspired me a lot to want to learn how to do this, keep up the great work :)

@SomeKindofHuman49 4 ай бұрын

C'est pas trop mon domaine mais ta vidéo m'a fait pensé à cet article : Data-driven prediction and control of extreme events in a chaotic flow (trouvable sur Scholar) J'espère que ça pourra t'aider à dompter le chaos !! Excellent travail en tout les cas !

@coreyaustin9706 3 ай бұрын

Brilliant work. Astounding.

@ponsmatheo1025 4 ай бұрын

"I don't want to dream of this pipes anymore" 😂 Love the video

@abraruralam3534 4 ай бұрын

30:55 "it seems to lack creativity." Which is a casual way of saying: "The AI is performing a gradient descent on its loss function. One of the limitations of gradient descent is that it will find the _local_ minima rather than the global minimum. Because after all, gradient descent is a greedy approach." Humans are "creative" because we can use a dynamic approach. We do often use a greedy approach too as greedy approach requires less planning, but we are also always questioning, "Maybe there's a better solution that's out of the box?" which is what dynamic approach does: it is always comparing if a unique set of subsolutions can beat the previously discovered optimal set of subsolutions, when combined with a new possible "frontier" of choices. In the case of turning the car around to forwards-facing, the AI skipped over the possibility of a long term improvement because it is using a greedy approach (it only sees the immediately-next best thing it could do, turning the car around at this moment will slow it down). A human driver, will, however, consider that turning the car around does increase it's speed for the rest of the track. They will therefore compare if a unique solution set consisting of the car turned forward-facing will possibly beat solution sets where the car is left backwards. E.g. Let's say the track had only 4 turns S1= {left, left, turn car around, right} VS. S2= {left, left, rigth, right} Both greedy approach and dynamic approach realize solution set S2 is faster as turning the car around takes too much time. The difference is, greedy will now NEVER try turning around ever again. If say even if the track had 8 turns rather than just 4. Dynamic will STILL experiment with turning around. S3= {left, left, turn car around, right, right, left, left, right} S4= {left, left, rigth, right, right, left, left, right} Turns out, solution set S3 is faster than S4 because turning the car around forwards makes it faster on the next 4 turns. Dynamic has found this solution because it always considers the entire solution space. Greedy will not find S3 because it is a superset of S1 and remember S1 was discarded by greedy so this superset optimal solution was never calculated by it. So yeah, this "What if?" curiosity in dynamic approach is what makes it capable of finding the global minimum.

@cocon16_PW 3 ай бұрын

I laughed out loud at 28:54 😂 the visuals at the end are absolutely stunning! The whole vid is a masterpiece.

@100andmore5 4 ай бұрын

This was an amazing video. I would LOVE to see a video where you try to get the ai to learn these creative tricks by themselves. Where the end is still possible to complete without using them.

@danielclayden1160 4 ай бұрын

I never comment on KZfaq but this was an absolutely amazing video! Content and script/editing were both perfect. Thank you for making it!!

@jammin023 4 ай бұрын

First video of yours I've seen, and it's an instant subscribe. I know nothing about Trackmania and I'm not much of a gamer generally, but this was absolutely fascinating, and the quality of your shots and editing is outstanding.

@maciodb 2 ай бұрын

wow, the work you have done is so impressive

@DanelonNicolas Ай бұрын

love this series of videos!! wanna see more!!

@rubenvangelder7359 4 ай бұрын

Yo man was just gonna say that you inspired me to dive deeper into reinforcement learning for my master thesis after seeing your amazing work. Keep it up!❤