Author Topic: When do you think A.I. will be able to break the Donkey Kong world record?  (Read 6429 times)

0 Members and 1 Guest are viewing this topic.

Offline homerwannabee

  • Spring Jumper
  • *
  • Posts: 1369
Now, I'm not talking about programming a computer with specific moves to beat the World Record.  What I'm talking about is a computer learning the game from scratch with only the idea of getting the highest score possible in it's program.  I've seen talk about how these A.I.'s are beating people in fighting games, and games like LoL, but what people fail to take into account with these type of games is that speed matters in all of these games.  If you are able to generate moves 3 or 4 times faster than your opponent, then of course you'll do really really well.

With a game like Donkey Kong though, speed means absolutely nothing.  You actually have to learn the game, but learning the game is not enough.  You need to get to the killscreen, and then relearn to score higher, and then relearn again to score even higher.

Also, another problem,  The major programmers actually caring enough to put A.I. to the test of Donkey Kong.  The major A.I. may not be programmed for that.  It seems rather random the games they choose.  So a lower level A.I. programmer may actually be the one who has the A.I. program to beat the Donkey Kong high score.

So based on everything I personally believe it's going to be around the year 2030 before an A.I. program actually beats the Donkey Kong Arcade/MAME world record.
"Perception forged in delusion and refined by pain"

-Ross Benzinger

"It's like we are able to play beautiful music out there, but no one can hear the instruments"

-Leon Shepard
DK3 World Record Holder DK Jr. Killscreener DKF Blogger Crazy Kong Killscreener Twitch TV  Streamer DK3 Repetitive Blue Screen

Online Scoundrl

  • Elite Member
  • *
  • Posts: 286
Any significant effort to program an AI to break the DK WR would result in that record being CRUSHED in short order. Auto Machine Learning is pretty powerful and still in its infancy.

This could easily happen before the KO6.

! No longer available






« Last Edit: January 21, 2018, 09:04:41 AM by Scoundrl »
Twitch TV  Streamer

Offline homerwannabee

  • Spring Jumper
  • *
  • Posts: 1369
Then explain this video?  19 months ago this A.I. can't even get 15k on the Atari 2600 Galaxian game.  The Atari 2600 game is WAY easier than the Arcade/MAME version.  https://www.youtube.com/watch?v=AVg_YIp09ps
"Perception forged in delusion and refined by pain"

-Ross Benzinger

"It's like we are able to play beautiful music out there, but no one can hear the instruments"

-Leon Shepard
DK3 World Record Holder DK Jr. Killscreener DKF Blogger Crazy Kong Killscreener Twitch TV  Streamer DK3 Repetitive Blue Screen

Offline homerwannabee

  • Spring Jumper
  • *
  • Posts: 1369
Also this is current A.I. playing Super Mario Bros live.  The computer has taken like 2 months, and it still hasn't completed level 1 of the game.

https://www.youtube.com/watch?v=QiXWqjNurQw
"Perception forged in delusion and refined by pain"

-Ross Benzinger

"It's like we are able to play beautiful music out there, but no one can hear the instruments"

-Leon Shepard
DK3 World Record Holder DK Jr. Killscreener DKF Blogger Crazy Kong Killscreener Twitch TV  Streamer DK3 Repetitive Blue Screen

Online Scoundrl

  • Elite Member
  • *
  • Posts: 286
That is FAR from state of the art AI, If google took on DK it would be solved by march.
Twitch TV  Streamer

Offline homerwannabee

  • Spring Jumper
  • *
  • Posts: 1369
That is FAR from state of the art AI, If google took on DK it would be solved by march.
Well that was another thing though I factored in.  If the top level A.I. programmers ignored something like this, than the next lower level programmers would tackle it, and by the looks of things the lower level A.I. programmers may take till 2030.
"Perception forged in delusion and refined by pain"

-Ross Benzinger

"It's like we are able to play beautiful music out there, but no one can hear the instruments"

-Leon Shepard
DK3 World Record Holder DK Jr. Killscreener DKF Blogger Crazy Kong Killscreener Twitch TV  Streamer DK3 Repetitive Blue Screen

Offline TheSunshineFund

  • Spring Jumper
  • *
  • Posts: 1109
  • Why not play some Kangaroo today?
the lower level A.I. programmers may take till 2030.

Maybe longer.  Shopping exclusively at Whole Foods and telling you how much the music and beer you like, actually sucks, takes up a lot of time.
« Last Edit: January 21, 2018, 01:49:52 PM by TheSunshineFund »
And as the cattlerack was cleared
The floor was brightly stained
And larger though we grew in size
Not a thing was gained
The fate as so ordained
Twitch TV  Streamer DK Killscreener DK Jr. Killscreener DKF Blogger Crazy Kong Killscreener

Offline marky_d

  • Global Moderator
  • Spring Jumper
  • *
  • Posts: 664
 ROFL
Twitch TV  Streamer DK Killscreener DK 1 Million Point Scorer DK Jr. Killscreener DK Jr. Arcade WR Holder DK 1.1 Million Point Scorer

Offline hooch66

  • Full Member
  • *
  • Posts: 86
    • Personal Blog
I don't see an AI as they are currently and without human coaching coming up with point pressing moves such as jumping near DK's foot on the girders anytime soon.
DKF Blogger

Offline danman123456

  • Spring Jumper
  • *
  • Posts: 711
    • Twitch TV Stream
That is FAR from state of the art AI, If google took on DK it would be solved by march.

Yeah if Alpha Zero was thrown at this it would likely be pretty quick all things considered. Probably a 1.3 million score within 3-6 months if that is what people wanted. Otherwise a KS would happen fairly quickly.

https://gaming.youtube.com/watch?v=QiXWqjNurQw
DK High : 1,059,700 (Lvl 22-1 KS!);
DK KS Speedrun : 1 Hr 16 Min 40 Sec - World Record!
DK Lvl 1 - 1: 12,400
Fix-It Felix Jr - 297,000 (World Record)
Fix-IT Felix Jr 1 Hour Limit - 177,000
Fix-It Felix Jr KS Speedrun - 1h33
Twitch TV  Streamer DKF Blogger Crazy Kong Killscreener DK Killscreener DK 1 Million Point Scorer

Offline KongTower

  • Jr. Member
  • *
  • Posts: 35
Re: When do you think A.I. will be able to break the Donkey Kong world record?
« Reply #10 on: February 06, 2018, 09:11:06 PM »
Once AI has the record for top score, it won't go back to a human.

The AI would need human help, obviously.  But these days, it's not about programming.  It's about having tons of raw processing power, tons of storage, and really fast memory and storage, and of course, being able to play faster than humanly possible (bypass real-time sync to display output) in order to reduce the training time.

Offline erockbrox

  • New Member
  • *
  • Posts: 7
Re: When do you think A.I. will be able to break the Donkey Kong world record?
« Reply #11 on: February 09, 2018, 01:13:47 AM »
The interesting thing about AI is that it is potentially limitless. While AI might be just getting started you really have to look at the concept in the long term. What is AI going to be like in 100 years? What will AI be like in 1,000 years? In 100,000 years?

These time spans might seems to be unreasonable to understand, but the fact is, the Earth will be here 100,000 years from now and who knows what the heck its going to be like to live on this future Earth.

I have always wondered what future AI will be like. Will it just be one big super computer which knows everything? Or will there be individual humanoid robots that are equipped with AI and each robot has their own unique personality that is shaped by their experiences much like us humans?

A future humanoid with AI could beat the Donkey Kong world record, but then again you would have to ask the robot "why do you want to beat the DK record?". A future robot with AI would have to have the desire to choose to beat the DK WR. It would have to be a personal choice made by the robot. We could program and tell a primitive AI bot to set the DK WR, but I don't think that we could tell an advanced AI what to do period.

If an advanced AI were to choose to try and beat the DK WR. I would imagine that it would go something like this:


1) Downloads MAME and Donkey Kong rom off the internet

2) Plays the Donkey Kong game on MAME with the game sped up by 1,000 times or more inside its own operating system

3) Plays the game several billions of times with in a short time span

4) Sits down at Donkey Kong arcade machine and basically preforms a "perfect" Donkey Kong game first try


Seriously that's what it's going to be like. It will be unreal.

« Last Edit: February 09, 2018, 01:21:16 AM by erockbrox »

Offline Zmick_tricK

  • Full Member
  • *
  • Posts: 54
Re: When do you think A.I. will be able to break the Donkey Kong world record?
« Reply #12 on: February 09, 2018, 05:02:18 PM »
If AI were truly "learning" without human coaching, would it not be just as likely to find game breaking glitches to manipulate score, including any possible avenues for arbitrary code execution, as it would to maximize "normal" gameplay?

I imagine watching one of these AI performed games live and then we get to the kill screen and... what the...??? He's not dying?! Because somewhere along the way, without anyone even noticing, some tricky movements were done earlier in the game to rewrite how the level counter is saved in memory and BAZOOM, kill screen averted!

This would of course depend on if such an exploit exists in the DK code, I'm just imagining how something like this could play out in the case that it did- and in the case as I'm sure it will with other games. Imagine AI that finds major glitches for speedruns because, taking the heuristics of human cognition out of the picture, the AI doesn't "know" what normal play looks like and doesn't automatically make any "assumptions" on how game mechanics work.

This, incidentally, is probably why it's taking so long for AI to do tasks that we consider simple. Without a heuristical base for decision making, it has to account for significantly more possibilities than a human ever would- including things that would seem ridiculous, like backtracking to another screen just to deliberately fall into a hole, hitting a specific pixel at a specific time on the game clock in the process, and then dying. You'd never know if that would cause a wrong warp because you'd never try it. However, AI might be able to find it by brute force.
« Last Edit: February 09, 2018, 05:04:14 PM by Zmick_tricK »
DKF Blogger

Offline KongTower

  • Jr. Member
  • *
  • Posts: 35
Re: When do you think A.I. will be able to break the Donkey Kong world record?
« Reply #13 on: February 09, 2018, 05:23:28 PM »
You're missing a critical thing.  Learning.  It needs data to learn from.

If you think you "figure out" every thing that you know how to do in the game, you didn't.

You used observation watching people better than you to learn much.

So will the AI.  Ahem.  Excuse me, Machine Learning.  http://knowledge.wharton.upenn.edu/article/democratization-ai-means-tech-innovation/

In any case, the probability of the AI finding exploits is not greater than the probability of the exploits existing.

DK is a pretty damn tight game.
« Last Edit: February 09, 2018, 05:25:23 PM by KongTower »

Offline KongTower

  • Jr. Member
  • *
  • Posts: 35
Re: When do you think A.I. will be able to break the Donkey Kong world record?
« Reply #14 on: February 09, 2018, 05:29:33 PM »
Let me tell a story.  When I was in college at Texas A&M back in 1990 for my AI class, I chose to write a self-playing Tetris.

It used what's known as "MINIMAX" algorithm.  It's basically an algorithm that sits on top of an evaluation for board layout saying "how good is this layout for continued play?"

It then "imagines" all the possible places the current piece could go, and puts the piece in the place that evaluates to the best score for how the level looks.

Any machine learning approach for Donkey Kong will make observations for how similar the screen looks to data for what the screen looked like preceding a known situation that led to death.

In other words, it will, all on it's own make neural connections based on what it "sees" about how "good" things look and try to come up with inputs known to get out of any tricky situations.

It's really not that complex given enough frames of evaluated donkey kong performances to compare what you're seeing at that moment.