One of the very interesting properties of Reinforcement Learning algorithms is that they allow learning without prior knowledge of the environment. However, when the agents use al...
Often when modeling structured domains, it is desirable to leverage information that is not naturally expressed as simply a label. Examples include knowledge about the evaluation ...
Most systems based on separation logic consider only restricted forms of implication or non-separating conjunction, as full support for these connectives requires a non-trivial no...
Aleksandar Nanevski, Josh Berdine, Viktor Vafeiadi...
The paper presents an approach to using structural descriptions, obtained through a human-robot tutoring dialogue, as labels for the visual object models a robot learns. The paper...
Geert-Jan M. Kruijff, John D. Kelleher, Gregor Ber...
This paper addresses the problem of learning from highly structured data. Speci cally, it describes a procedure, called decomposition, that allows a learner to access automatically...