An important feature of many problem domains in machine learning is their geometry. For example, adjacency relationships, symmetries, and Cartesian coordinates are essential to an...
We propose a simple approach to combining first-order logic and probabilistic graphical models in a single representation. A Markov logic network (MLN) is a first-order knowledge b...
This paper presents a method to induce relational concepts with neural networks using the inductive logic programming system LINUS. Some first-order inductive learning tasks taken...
Rodrigo Basilio, Gerson Zaverucha, Artur S. d'Avil...
We present cdec, an open source framework for decoding, aligning with, and training a number of statistical machine translation models, including word-based models, phrase-based m...
Chris Dyer, Adam Lopez, Juri Ganitkevitch, Jonatha...
We present a new approach to reinforcement learning in which the policies considered by the learning process are constrained by hierarchies of partially specified machines. This ...