Answering the “why” in answer set programming – A survey of explanation approaches

@article{Fandinno2019AnsweringT,
  title={Answering the “why” in answer set programming – A survey of explanation approaches},
  author={Jorge Fandinno and Claudia Schulz},
  journal={Theory and Practice of Logic Programming},
  year={2019},
  volume={19},
  pages={114 - 203}
}
Abstract Artificial intelligence (AI) approaches to problem-solving and decision-making are becoming more and more complex, leading to a decrease in the understandability of solutions. The European Union’s new General Data Protection Regulation tries to tackle this problem by stipulating a “right to explanation” for decisions made by AI systems. One of the AI paradigms that may be affected by this new regulation is answer set programming (ASP). Thanks to the emergence of efficient solvers, ASP… 
Justifications for Goal-Directed Constraint Answer Set Programming
Ethical and legal concerns make it necessary for programs that may directly influence the life of people (via, e.g., legal or health counseling) to justify in human-understandable terms the advice
Explainable ASP
TLDR
A given answer set is used as guidance for a simplified top-down procedure of answer set semantics developed by Satoh and Iwayama to provide not only an explanation for the derivation (or non-derivation) of the atoms, but also an explanation of the consistency of the whole answer set itself.
Towards Comprehensible ASP Reasoning by Means of Abstraction
TLDR
This work highlights the potential of the recently introduced abstraction concept in ASP towards tackling this problem, by focusing on the relevant details of the reasoning task that can allow for better human comprehensibility.
Towards Generating Explanations for ASP-Based Link Analysis using Declarative Program Transformations
TLDR
This paper investigates the problem of link analysis, specifically link prediction and anomalous link discovery in social networks using the declarative method of Answer set programming (ASP), and proposes a novel method for generating explanations - as offline justifications - usingDeclarative program transformations.
Machine Reasoning Explainability
TLDR
This paper aims to provide a selective overview of MR explainability techniques and studies in hopes that insights from this long track of research will complement well the current XAI landscape.
Abstraction for Zooming-In to Unsolvability Reasons of Grid-Cell Problems
TLDR
This paper focuses on obtaining the reasons for unsolvability of problems on grids, and shows the possibility to automatically achieve human-like abstractions that distinguish only the relevant part of the grid.
Explaining Non-Acceptability in Abstract Argumentation
TLDR
The resulting system is thus able to deliver explanations also in the case of non-acceptance and is provided a first empirical study that shows the feasibility of this approach.
Introduction to the TPLP Special Issue on User-oriented Logic Programming and Reasoning Paradigms
With the rise of machine learning, and more recently the overwhelming interest in deep learning, knowledge representation and reasoning (KRR) approaches struggle to maintain their position within the
A System for Explainable Answer Set Programming
We present xclingo, a tool for generating explanations from ASP programs annotated with text and labels. These annotations allow tracing the application of rules or the atoms derived by them. The
Argumentative XAI: A Survey
TLDR
This survey overviews the literature focusing on different types of explanation, different models with which argumentation-based explanations are deployed, different forms of delivery, and different argumentation frameworks they use, and lays out a roadmap for future work.
...
...

References

SHOWING 1-10 OF 117 REFERENCES
ASPeRiX, a first-order forward chaining approach for answer set computing*
TLDR
The project ASPeRiX is presented as a first-order forward chaining approach for Answer Set Computing, among the first to introduce an approach of answer set computing that escapes the preliminary phase of rule instantiation by integrating it in the search process.
Justifications and Blocking Sets in a Rule-Based Answer Set Computation
TLDR
The paper addresses the question in a rule-based answer set computation: the search algorithm does not guess on the truth or falsity of an atom but on the application or non application of a non monotonic rule.
A Meta-Programming Technique for Debugging Answer-Set Programs
TLDR
This paper introduces a meta-programming technique for debugging in ASP that respects the declarative nature of ASP, and the capabilities of the system can easily be extended to incorporate differing debugging features.
Stepwise debugging of answer-set programs*
TLDR
A framework of computations for answer-set programs is developed on an abstract ASP language that is sufficiently general to capture different solver languages and makes use of abstract constraints as an established abstraction for popular language constructs such as aggregates.
Formalization of psychological knowledge in answer set programming and its application
TLDR
This paper develops an ASP-based formalization of the mechanics of Short-Term Memory and shows that the approach can have rather immediate practical uses by demonstrating an application of the formalization to the task of predicting a user's interaction with a graphical interface.
Debugging ASP using ILP
TLDR
Inductive logic programming (ILP) along with examples of (un)desirable properties of answer sets can be used to revise the original program semi-automatically so that it satisfies the stated properties, in effect providing debugging-by-example for programs under answer set semantics.
Unifying Justifications and Debugging for Answer-Set Programs
TLDR
It is shown that the provenance approach generalizes the debugging one, since any error has a counterpart provenance but not the other way around.
Justifying answer sets using argumentation
TLDR
It is shown how argumentation theory can help to explain why a literal is or is not contained in a given answer set by defining two justification methods, both of which make use of the correspondence between answer sets of a logic program and stable extensions of the assumption-based argumentation (ABA) framework constructed from the same logic program.
ASSAT: computing answer sets of a logic program by SAT solvers
Epistemological Problems of Artificial Intelligence
...
...