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 "enisoc" backwards in the box below (case sensitive):

Archive

Show me a random blog post
 2021 

Jan 2021

Christmas (2020) is over
 2020 
▼ show ▼
 2019 
▼ show ▼
 2018 
▼ show ▼
 2017 
▼ show ▼
 2016 
▼ show ▼
 2015 
▼ show ▼
 2014 
▼ show ▼
 2013 
▼ show ▼
 2012 
▼ show ▼

Tags

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

Archive

Show me a random blog post
▼ show ▼
© Matthew Scroggs 2012–2021