mscroggs.co.uk
mscroggs.co.uk

subscribe

Blog

 2019-12-27 
In tonight's Royal Institution Christmas lecture, Hannah Fry and Matt Parker demonstrated how machine learning works using MENACE.
The copy of MENACE that appeared in the lecture was build and trained by me. During the training, I logged all the moved made by MENACE and the humans playing against them, and using this data I have created some visualisations of the machine's learning.
First up, here's a visualisation of the likelihood of MENACE choosing different moves as they play games. The thickness of each arrow represented the number of beads in the box corresponding to that move, so thicker arrows represent more likely moves.
The likelihood that MENACE will play each move.
There's an awful lot of arrows in this diagram, so it's clearer if we just visualise a few boxes. This animation shows how the number of beads in the first box changes over time.
The beads in the first box.
You can see that MENACE learnt that they should always play in the centre first, an ends up with a large number of green beads and almost none of the other colours. The following animations show the number of beads changing in some other boxes.
MENACE learns that the top left is a good move.
MENACE learns that the middle right is a good move.
MENACE is very likely to draw from this position so learns that almost all the possible moves are good moves.
The numbers in these change less often, as they are not used in every game: they are only used when the game reached the positions shown on the boxes.
We can visualise MENACE's learning progress by plotting how the number of beads in the first box changes over time.
The number of beads in MENACE's first box.
Alternatively, we could plot how the number of wins, loses and draws changes over time or view this as an animated bar chart.
The number of games MENACE wins, loses and draws.
The number of games MENACE has won, lost and drawn.
If you have any ideas for other interesting ways to present this data, let me know in the comments below.

Similar posts

Building MENACEs for other games
MENACE at Manchester Science Festival
MENACE
MENACE in fiction

Comments

Comments in green were written by me. Comments in blue were not written by me.
@(anonymous): Have you been refreshing the page? Every time you refresh it resets MENACE to before it has learnt anything.

It takes around 80 games for MENACE to learn against the perfect AI. So it could be you've not left it playing for long enough? (Try turning the speed up to watch MENACE get better.)
Matthew
                 Reply
I have played around menace a bit and frankly it doesnt seem to be learning i occasionally play with it and it draws but againt the perfect ai you dont see as many draws, the perfect ai wins alot more
(anonymous)
                 Reply
@Colin: You can set MENACE playing against MENACE2 (MENACE that plays second) on the interactive MENACE. MENACE2's starting numbers of beads and incentives may need some tweaking to give it a chance though; I've been meaning to look into this in more detail at some point...
Matthew
                 Reply
Idle pondering (and something you may have covered elsewhere): what's the evolution as MENACE plays against itself? (Assuming MENACE can play both sides.)
Colin
                 Reply
 Add a Comment 


I will only use your email address to reply to your comment (if a reply is needed).

Allowed HTML tags: <br> <a> <small> <b> <i> <s> <sup> <sub> <u> <spoiler> <ul> <ol> <li>
To prove you are not a spam bot, please type "rotcev" backwards in the box below (case sensitive):

Archive

Show me a random blog post
 2020 

Jul 2020

Happy τ+e-6 Approximation Day!

May 2020

A surprising fact about quadrilaterals
Interesting tautologies

Mar 2020

Log-scaled axes

Feb 2020

PhD thesis, chapter ∞
PhD thesis, chapter 5
PhD thesis, chapter 4
PhD thesis, chapter 3
Inverting a matrix
PhD thesis, chapter 2

Jan 2020

PhD thesis, chapter 1
Gaussian elimination
Matrix multiplication
Christmas (2019) is over
 2019 
▼ show ▼
 2018 
▼ show ▼
 2017 
▼ show ▼
 2016 
▼ show ▼
 2015 
▼ show ▼
 2014 
▼ show ▼
 2013 
▼ show ▼
 2012 
▼ show ▼

Tags

dataset curvature the aperiodical wave scattering ternary interpolation geometry javascript folding tube maps frobel sound hexapawn christmas chess cambridge rhombicuboctahedron radio 4 computational complexity asteroids london menace electromagnetic field pi phd reddit simultaneous equations matrix multiplication craft hats graphs matrices football platonic solids weather station pac-man arithmetic sorting boundary element methods a gamut of games estimation ucl go inline code matrix of minors propositional calculus inverse matrices trigonometry flexagons world cup squares data visualisation sport logs programming pythagoras books error bars people maths fractals mathslogicbot quadrilaterals php map projections mathsjam misleading statistics oeis finite element method game of life manchester sobolev spaces pi approximation day speed games binary approximation wool christmas card signorini conditions chalkdust magazine folding paper cross stitch tmip weak imposition triangles chebyshev london underground countdown braiding accuracy statistics final fantasy manchester science festival machine learning tennis probability light hannah fry stickers convergence reuleaux polygons captain scarlet golden spiral royal baby royal institution golden ratio game show probability nine men's morris draughts graph theory big internet math-off news determinants noughts and crosses mathsteroids advent calendar realhats coins national lottery bubble bobble talking maths in public matrix of cofactors dragon curves gaussian elimination gerry anderson data plastic ratio dates video games european cup harriss spiral python pizza cutting rugby bempp martin gardner matt parker latex preconditioning geogebra numerical analysis palindromes twitter exponential growth bodmas polynomials raspberry pi puzzles logic

Archive

Show me a random blog post
▼ show ▼
© Matthew Scroggs 2012–2020