Louise is a Meta-Interpretive Learning (MIL) system. MIL is like a second-order Prolog where first-order programs are learned from higher-order programs by Resolution. There's a long thread of literature on MIL going back to 2014 but it now seems we're starting to move towards applications, e.g. I'm doing a post-doc where I use MIL to learn autonomous behaviours for an agent that must guide a mobile robot in survey missions. Other colleagues are working on applications in biology. We're going slowly because there's very few of us but MIL is a powerful technique that extends the soundness and completeness of SLD-Resolution to induction, so I'm hopeful that good things will happen with a bit of elbow grease and a bit of patience.
https://github.com/stassa/louise
Louise is a Meta-Interpretive Learning (MIL) system. MIL is like a second-order Prolog where first-order programs are learned from higher-order programs by Resolution. There's a long thread of literature on MIL going back to 2014 but it now seems we're starting to move towards applications, e.g. I'm doing a post-doc where I use MIL to learn autonomous behaviours for an agent that must guide a mobile robot in survey missions. Other colleagues are working on applications in biology. We're going slowly because there's very few of us but MIL is a powerful technique that extends the soundness and completeness of SLD-Resolution to induction, so I'm hopeful that good things will happen with a bit of elbow grease and a bit of patience.