The source code for chapter 3 of my thesis: "Aritificial Understanding"

The larger part of the code is a direct copy of https://github.com/RichardEvans/apperception. All additions concern the implementation of 'causal judgements', that harness the full expressive power of geometric logic.

Installation instructions

You need to have installed Haskell and Clingo (version 4.5 or above).

1. To install Haskell:

   • go to https://www.haskell.org/downloads/

2. To install Clingo (version 4.5 or above):

   • go to https://potassco.org/clingo/

Compilation instructions

Once you have Haskell and Clingo installed, just run (from the root directory):

• cd code
• cabal update
• cabal configure
• cabal new-build
• cabal install
• cd ..

Examples

Once the system is installed (see above), you are ready to try some examples developed for the original Apperception Engine by Richard Evans:

To run these examples, make sure you are in the root directory called geometric_apperception.

A single sensor oscillating between on and off:

• ~/.cabal/bin/mysolve misc predict_1.lp

Two sensors, one oscillates between on and off, while the other has the same reading throughout:

• ~/.cabal/bin/mysolve misc predict_2.lp

Exogenous action:

• ~/.cabal/bin/mysolve misc exog_1.lp

Searching through templates of increasing complexity, looking for a unified interpretation:

• ~/.cabal/bin/mysolve eca-general predict_eca_245_b3.lp

More Complex Examples

See https://github.com/RichardEvans/apperception for a list of more complex examples for the original system.

Now to make use of existential quantification in the head of judgements you can try the following examples:

A simple counting sequence:

• ~/.cabal/bin/mysolve cj heating_2.lp

A simple example in which conjunction of atoms in the head is helpful:

• ~/.cabal/bin/mysolve cj conjunction_1.lp

The final example of chapter 3 (this takes about 2 hours) :

• ~/.cabal/bin/mysolve cj heating_1.lp

In general, mysolve can be run with any file in the data directory. All examples added by me are in the data/causal_judgement folder.

Understanding the output of the solve process

When solve is run, it produces...

• the theory θ = (φ, I, R, C) composed of...
  • the initial conditions (I)
  • the rules (R)
  • the constraints (C)
• the trace (τ(θ))
• statistics: the cost of the interpretation θ
• accuracy: whether or not all the predicted sensor readings match the hidden readings

To generate a latex-readable description of the output:

• set flag_output_latex = True in Interpretation.hs
• recompile: scripts/compile_solve.sh
• run again

Data generation

The data is already provided in the data/ folder.

But if you want to regenerate it:

• scripts/compile_all.sh
• cd code
• ./eca all
• ./sw all
• ./music all
• ./rhythm all