A.I. Learns - Walk or Run in the Rain? (150.000+ trials)

#artificialintelligence #educational #machinelearning

Is it better to walk or run the rain?

Description
------------------
In this video, we train an A.I. to help us answer this age old question.
Although there have been many attempts in the past to resolve this debate once and for all, we still don’t have a definitive answer. But don’t worry. When humans fail to give an answer. There is always A.I. After extensive research I found out that in general the studies and experiments conducted are almost evenly split between supporters of Walking in the rain and supporters running in the rain. Watch the video to find out the answer our A.I. has given to this query.

About the Channel
------------------------------
Artificial Intelligence is fun. On this channel, we train AIs to answer everyday questions and upload the simulations online. The questions come directly from our community,so if you have something that needs answering through Artificial Intelligence entities, then here's the channel to be :)

P.S. This is also an investment for when our future AI overlords take over.

~ Giorgos Demosthenous

Stay up to date
-------------------------

Music
----------
Golden by Mona Wonderlick @monawonderlick

Stock footage
-----------------------

Thumbnail photo by VisionPic .net from Pexels

Some Resources
---------------------------

[jgcornell]

This is great, but really relies on the accuracy of the rain fluid simulation

[willguggn]

If there was an optimal running speed in-between walking and running the AI could never find it because it just switches between running and walking.

So you can only conclude that the running speed you chose is better than the walking speed you chose.

[Protectorio]

In my opinion the best choice is to use an umbrella.

[justinwhite]

.. There is a mathematically definitive answer. The answer is to run. Sideways hitting drops are a function of distance (which is the same whether you run or walk) and downward falling drops are a function of time.

You will get wet faster by running, but you will be less wet by the time you reach your destination than if you had walked.

[MetrotomX]

How detailed is the rain model here? Does it actually simulate based upon individual droplets colliding with the entity, or is the wetness just a simple incremented value based upon time spent in rain and rain intensity?


Is the wind direction/intensity randomized between trials? Does the wind direction/intensity change within a trial?


Any idea why the AI doesn't choose to run 100% of the time?

[harishthirunagari]

Make an other video with variables like the direction of the rain, the size of the water droplets and see what the ai does in each scenario.

[IronMan-wzdx]

What if it has to monitor like battery levels?

[justinwhite]

You failed to mention how the wetness value is calculated. Is it based on particles hit or something else?

[Hyrtsi]

What if the person was really tall? If the training length isn't that many days (like it seems) maybe you could do a bunch of different scenarios as a follow-up :) nice work anyways!

[KalimbaRlz]

it would be nice to have the code you have run, thanks

[fackarov]

the quantum-teleportation is the way to go

[wickedgentleman]

If it is deciding wether to run or walk, couldnt there be just a pattern to the rain drops of which the AI learned to follow? Like fiding the exact spot and time to get the minimal wetness running to the spot and walking cause you can see it found an optimal speed but it still run stuttering, avoiding the rain

[Cd.Morpheus]

All this time I thought it was better to walk 😭

[Abishek_Muthian]

That was interesting, link to source papers and perhaps your own repo would add more value to the video.

[AdzFFS]

I laughed so hard when i saw the fat agent

[yossishm]

What about the chance of falling into a puddle?

[bleuebloom]

umm, couldn't u have just run the simulation 100X at all 100 walk/run ratios?

[friday]

Thats great and all but it shows that there must be some sort of flaw in your rain simulation because when mythbusters did it in real life they got the opposite result