Inductive logic programming at 30

Inductive logic programming (ILP) is a form of logic-based machine learning. The goal is to induce a hypothesis (a logic program) that generalises given training examples and background knowledge. As ILP turns 30, we review the last decade of research. We focus on (i) new meta-level search methods, (ii) techniques for learning recursive programs, (iii) new approaches for predicate invention, and (iv) the use of different technologies. We conclude by discussing current limitations of ILP and directions for future research.

Generating contrastive explanations for inductive logic programming based on a near miss approach

In recent research, human-understandable explanations of machine learning models have received a lot of attention. Often explanations are given in form of model simplifications or visualizations. However, as shown in cognitive science as well as in early AI research, concept understanding can also be improved by the alignment of a given instance for a concept with a similar counterexample. Contrasting a given instance with a structurally similar example which does not belong to the concept highlights what characteristics are necessary for concept membership. Such near misses have been proposed by Winston (Learning structural descriptions from examples, 1970) as efficient guidance for learning in relational domains. We introduce an explanation generation algorithm for relational concepts learned with Inductive Logic Programming (GeNME). The algorithm identifies near miss examples from a given set of instances and ranks these examples by their degree of closeness to a specific positive instance. A modified rule which covers the near miss but not the original instance is given as an explanation. We illustrate GeNME with the well-known family domain consisting of kinship relations, the visual relational Winston arches domain, and a real-world domain dealing with file management. We also present a psychological experiment comparing human preferences of rule-based, example-based, and near miss explanations in the family and the arches domains.

Lifting Symmetry Breaking Constraints with Inductive Logic Programming

Efficient omission of symmetric solution candidates is essential for combinatorial problem solving. Most of the existing approaches are instance-specific and focus on the automatic computation of Symmetry Breaking Constraints (SBCs) for each given problem instance. However, the application of such approaches to large-scale instances or advanced problem encodings might be problematic. Moreover, the computed SBCs are propositional and, therefore, can neither be meaningfully interpreted nor transferred to other instances. To overcome these limitations, we introduce a new model-oriented approach for Answer Set Programming that lifts the SBCs of small problem instances into a set of interpretable first-order constraints using the Inductive Logic Programming paradigm. Experiments demonstrate the ability of our framework to learn general constraints from instance-specific SBCs for a collection of combinatorial problems. The obtained results indicate that our approach significantly outperforms a state-of-the-art instance-specific method as well as the direct application of a solver.

Inductive logic programming applied for knowledge representation in computer music/ Programação lógica indutiva aplicada para representação do conhecimento em música computacional

In Computer Music, the knowledge representation process is an essential element for the development of systems. Methods have been applied to provide the computer with the ability to generate conclusions based on previously established experience and definitions. In this sense, Inductive Logic Programming presents itself as a research field that incorporates concepts of Logic Programming and Machine Learning, its declarative character allows musical knowledge to be presented to non-specialist users in a naturally understandable way. The present work performs a systematic review based on approaches that use Inductive Logic Programming in the representation of musical knowledge. Questions that these studies seek to address were raised, as well as identifying characteristic aspects related to their application.

Apperception

This paper describes a neuro-symbolic system for distilling interpretable logical theories out of streams of raw, unprocessed sensory experience. We combine a binary neural network, that maps raw sensory input to concepts, with an inductive logic programming system, that combines concepts into declarative rules. Both the inductive logic programming system and the binary neural network are encoded as logic programs, so the weights of the neural network and the declarative rules of the theory can be solved jointly as a single SAT problem. This way, we are able to jointly learn how to perceive (mapping raw sensory information to concepts) and apperceive (combining concepts into declarative rules). We apply our system, the Apperception Engine, to the Sokoban domain. Given a sequence of noisy pixel images, the system has to construct objects that persist over time, extract attributes that change over time, and induce rules explaining how the attributes change over time. We compare our system with a neural network baseline, and show that the baseline is significantly outperformed by the Apperception Engine.

Interactive Learning with Mutual Explanations in Relational Domains

With the growing number of applications of machine learning in complex real-world domains machine learning research has to meet new requirements to deal with the imperfections of real world data and the legal as well as ethical obligations to make classifier decisions transparent and comprehensible. In this contribution, arguments for interpretable and interactive approaches to machine learning are presented. It is argued that visual explanations are often not expressive enough to grasp critical information which relies on relations between different aspects or sub-concepts. Consequently, inductive logic programming (ILP) and the generation of verbal explanations from Prolog rules is advocated. Interactive learning in the context of ILP is illustrated with the Dare2Del system which helps users to manage their digital clutter. It is shown that verbal explanations overcome the explanatory one-way street from AI system to user. Interactive learning with mutual explanations allows the learning system to take into account not only class corrections but also corrections of explanations to guide learning. We propose mutual explanations as a building-block for human-like computing and an important ingredient for human AI partnership.

Logic-based Robotics

This chapter explores methods for combining symbolic and sub-symbolic reasoning and learning systems to take advantage of the strengths of each approach in challenging tasks in robotics. In perception, Inductive Logic Programming (ILP) can be used to learn descriptions of classes of objects and to find relations between objects. Examples are given of perception for robots in urban search and rescue. We also describe systems for learning plans and behaviours for robots. Relational learning is used to acquire abstract model of robot actions that are then used to constrain sub-symbolic learning for low-level control. Models can be variously expressed in the classical STRIPS representation or as qualitative models. A STRIPS-like model is acquired by a robot that learns to use tools and also designs new tools. A qualitative model is constructed by a robot that learns to traverse uneven terrain in urban search and rescue. The model is refined by reinforcement learning.