This is a past event
A conceptual presentation about implementing formal argumentation in Haskell, a functional programming language
Outline:
The departmental presentation will discuss the Haskell implementation on a much more conceptual level with more detailed discussion of the Haskell code. The session could be interactive session, combining slides with code and a whiteboard, addressing questions about the code and Haskell.
For a bio for Bas, see:
http://www.abdn.ac.uk/cops/events/4917/
Reading:
- There's the previously mentioned "Towards a framework for the implementation and verification of translations between argumentation models": http://www.cs.nott.ac.uk/~bmv/Papers/programming_dung_ifl_post.pdf
This discusses all three implementations (Carneades, Dung's AFs and the translation). It also shortly talks about the formalisation of Dung's AFs into a theorem prover.
- You can find the code for all three implementations on my website: http://www.cs.nott.ac.uk/~bmv/ or my github: https://github.com/nebasuke
What might be of interest is the actual formalisation of Dung into Agda: http://www.cs.nott.ac.uk/~bmv/Code/AF2.agda (You need to view this with encoding to Unicode!)
- There's "Tools for the implementation of argumentation models" which discusses some simple techniques like open source repositories and other things like literate programming which can help reuse of implementations. See: http://www.cs.nott.ac.uk/~bmv/Papers/iccsw2013_revised.pdf
- If someone actually wants to see the actual definition of the translation and the corresponding algorithm they can of course look at "Relating Carneades with abstract argumentation via the ASPIC+ framework for structured argumentation". See: http://www.cs.nott.ac.uk/~bmv/Papers/tARCrelating_preprint.pdf
This is probably a bit too much, but even skimming the paper does give a good impression of why it might be hard to actually implement a translation.
- Speaker
- Bas van Gijzel
- Hosted by
- Adam Wyner
- Venue
- MT 203