Explanatory guided learning

This is the code that accompanies the paper “Machine Guides, Human Supervises: Interactive Learning with Global Explanations”

Dependencies

The requirements.txt contains the Python dependencies and they can be installed using:

pip install -r requirements.txt

Experiments

To run the experiments, use the main.py script. Type python main.py --help for the list of options.

For instance, to run 10 folds of 100 iterations for the synthetic experiment with XGL(rules) and the competitors use:

python main.py --experiments synthetic --strategies al_dw al_lc random sq_random xgl_rules_simple_tree

The code will save all results in the results directory in pickle format.

Plots

To draw the plots, use the draw.py script. Type python draw.py --help for the list of options.

For example, to draw the plots for XGL(rules) and the competitors run:

python draw.py --folder <name_of_folder_containing_pickles> --strategies al_dw al_lc random sq_random xgl_rules_simple_tree

To draw the plots for XGL(rules) for different values of the parameter θ run:

python draw.py --folder <name_of_folder_containing_pickles> --strategies xgl_rules_simple_tree --thetas_rules 100.0 10.0 1.0