Reinforcement learning (RL) is commonly used to learn reward-optimizing policies. However, RL policies are not always trained with ethical behavior in mind, which can lead an agent to violate social or legal norms in pursuit of its goal. Retraining agents with additional norms is not always feasible, especially in complex stochastic environments. To mitigate this issue, we present a probabilistic policy fixing framework that adapts norm-agnostic policies online. Using Answer Set Programming (ASP), we generate policy fixes that minimize deviations from the RL policy while optimizing for norm adherence against a set of sampled worlds. Based on the Rule of Three and Hoeffding's inequality, we provide guarantees that fixed policies are near optimal, given a specified level of confidence.

Learning rules plays a crucial role in deep learning, particularly in explainable artificial intelligence and enhancing the reasoning capabilities of large language models. While existing rule learning methods are primarily designed for symbolic data, learning rules from image data without supporting image labels and automatically inventing predicates remains a challenge. In this paper, we tackle these inductive rule learning problems from images with a framework called γILP, which provides a fully differentiable pipeline from image constant substitution to rule structure induction. Extensive experiments demonstrate that γILP achieves strong performance not only on classical symbolic relational datasets but also on relational image data and pure image datasets, such as Kandinsky patterns.

Answer Set Programming (ASP) is a popular declarative reasoning and problem solving approach in symbolic AI. Its rule-based formalism makes it inherently attractive for explainable and interpretive reasoning, which is gaining increasing importance with the surge of Explainable AI (XAI). A number of explanation approaches and tools for ASP have been developed, which often tackle specific explanatory settings and may not cover all scenarios that ASP users might encounter. In this survey, we provide, guided by an XAI perspective, an overview of types of ASP explanations in connection with user questions for explanation, and describe how their coverage by current theory and tools in ASP. Furthermore, we pinpoint gaps in existing ASP explanations approaches and identify research directions for future work.

Visual Question Answering (VQA) is the task of answering questions about images, requiring the integration of multimodal input and reasoning. Modular approaches that incorporate logic-based representations into the reasoning component offer clear advantages over end-to-end trained systems, particularly in terms of interpretability. However, adapting or extending these representations when task requirements change can place a significant burden on developers. To address this challenge, we present an approach for distilling rules from Large Language Models (LLMs). Our method prompts an LLM to extend an initial VQA reasoning theory, expressed as an answer-set program, to meet new requirements of the task. Examples from VQA datasets guide the LLM, validate the results, and help correct erroneous rules by leveraging feedback from the ASP solver. We demonstrate that our approach is effective across diverse VQA datasets. Notably, only a few examples are needed to elicit correct rules from LLMs. Our experiments suggest that rule distillation from LLMs is a promising alternative to traditional data-driven rule learning approaches. Under consideration in Theory and Practice of Logic Programming (TPLP).

Reinforcement Learning (RL) enables autonomous agents to learn policies from experience, but realistic problems often involve enormous state spaces, making learning and generalisation challenging. Abstraction and approximation are therefore essential. Relational Reinforcement Learning (RRL) offers a way to reason about objects and their relations, and the CARCASS framework by Martijn van Otterlo demonstrates how logical representations can model Markov Decision Processes (MDPs) in first-order domains. Originally implemented in Prolog, CARCASS leverages domain knowledge to create powerful abstractions. We explore Answer-Set Programming (ASP), which is a rich and, contrary to Prolog, fully declarative modelling language, to realise CARCASS abstractions. We evaluate our ASP-based implementation in case studies of two domains, viz. Blocks World and Minigrid. Our results indicate that CARCASS with ASP provides a promising approach to constructing abstractions for RL, especially when domain knowledge is available.

Integration of Answer Set Programming (ASP) with neural networks has emerged as a promising tool in Neuro-symbolic AI. While existing approaches extend the capabilities of ASP to real world domains, their reasoning pipelines depend on classical solvers, which is a bottleneck for scalability. To tackle this problem, we propose a new method to compute stable models, called decision-propagation (DProp), which alternates falsity decisions and truth propagations. Successful DProp computations are shown to capture the stable model semantics. We then develop Neural DProp (NDProp), a differentiable extension of DProp with neural computation for decisions and fuzzy evaluation for propagations. We evaluate the capabilities of NDProp for learning decision heuristics as well as neuro-symbolic integration, and compare it with existing neuro-symbolic approaches. The results show that NDProp can learn to efficiently compute stable models, and it improves accuracy and scalability on neuro-symbolic benchmarks.

Explainability remains a key issue in reinforcement learning (RL). Distilling an interpretable policy from an agent trained in a complex environment is particularly challenging when the action space is continuous. We introduce ORCAID, a novel method for extracting interpretable rule-based policies from RL agents operating in mixed continuous-discrete environments with continuous action spaces. Our main contribution is an efficient oblique decision tree training algorithm that partitions the state space by hyperplanes and fits local linear models. The key idea lies in a three-stage split search: efficient random initialization, local refinement, and backward elimination. Finally, adjacent leaves are merged to yield a concise set of interpretable rules describing a given deep RL policy. We evaluate ORCAID across multiple RL environments, demonstrating that the extracted rule-based policies maintain strong performance with a low number of parameters and can even be used to improve the performance of the original deep RL policy.

This extended abstract presents GS-VQA, a neurosymbolic system for zero-shot Visual Question Answering (VQA). GS-VQA constructs symbolic, question-conditioned scene graphs from real-world images using zero-shot vision models guided by large language models. These graphs are effectively knowledge graphs that can be used for logic-based inference using Answer-Set Programming (ASP). The system enables question answering via symbolic inference and can generate logical explanation traces using xclingo. Evaluations on the GQA benchmark demonstrate the method's transparency and diagnostic power despite modest accuracy in comparison to state of the art neural systems.

Visual Question Answering (VQA) is a challenging problem that requires to process multimodal input. Answer-Set Programming (ASP) has shown great potential in this regard to add interpretability and explainability to modular VQA architectures. In this work, we address the problem of how to integrate ASP with modules for vision and natural language processing to solve a new and demanding VQA variant that is concerned with images of graphs (not graphs in symbolic form). Images containing graph-based structures are an ubiquitous and popular form of visualisation. Here, we deal with the particular problem of graphs inspired by transit networks, and we introduce a novel dataset that amends an existing one by adding images of graphs that resemble metro lines. Our modular neuro-symbolic approach combines optical graph recognition for graph parsing, a pretrained optical character recognition neural network for parsing labels, Large Language Models (LLMs) for language processing, and ASP for reasoning. This method serves as a first baseline and achieves an overall average accuracy of 73% on the dataset. Our evaluation provides further evidence of the potential of modular neuro-symbolic systems, in particular with pretrained models that do not involve any further training and logic programming for reasoning, to solve complex VQA tasks.

In recent years, quantitative complexity over semirings has been intensively investigated. An important problem in this context is to connect computational complexity with logical expressiveness. In this paper we improve on the model of Semiring Turing Machines (distinct from so called weighted Turing machines) introduced by Eiter & Kiesel (Semiring Reasoning Frameworks in AI and Their Computational Complexity, J. Artif. Intell. Res., 2023). Our central result is a Fagin-style theorem for a new quantitative complexity class using a suitable weighted logical formalism. We show that the quantitative complexity class that we call ℛ, where ℛ is a commutative semiring, can be captured using a version of weighted existential second-order logic that allows for predicates interpreted as semiring-annotated relations. This result provides a precise logical characterization of the power series that form the class ℛ. We also give the exact relation between Eiter & Kiesel's version of NP, called ℛ, and the class ℛ. Incidentally, we are able to recapture all the complexity results by Eiter & Kiesel (2023) in our new model, connecting a quantitative version of NP to various counting complexity classes.

Visual Question Answering (VQA) is the task of answering natural language questions about images, which is a challenge for AI systems. To enhance adaptability and reduce training overhead, we address VQA in a zero-shot setting by leveraging pre-trained neural modules without additional fine-tuning. Our proposed hybrid neurosymbolic framework, whose capabilities are demonstrated on the challenging GQA dataset, integrates neural and symbolic components through logic-based reasoning via Answer-Set Programming. Specifically, our pipeline employs large language models for semantic parsing of input questions, followed by the generation of a scene graph that captures relevant visual content. Interpretable rules then operate on the symbolic representations of both the question and the scene graph to derive an answer. Our framework provides a key advantage: it enables full transparency into the reasoning process. Using an existing explanation tool, we illustrate how our method fosters trust by making decisions interpretable and facilitates error analysis when predictions are incorrect. Beyond explaining its own reasoning, our framework can also explain answers from more opaque models by integrating their answers into our system, enabling broader interpretability in VQA.(1)

Reinforcement learning is a widely used approach for training an agent to maximize rewards in a given environment. Action policies learned with this technique see a broad range of applications in practical areas like games, healthcare, robotics, or autonomous driving. However, enforcing ethical behavior or norms based on deontic constraints that the agent should adhere to during policy execution remains a complex challenge. Especially constraints that emerge after the training can necessitate to redo policy learning, which can be costly and, more critically, time-intense. In order to mitigate this problem, we present a framework for policy fixing in case of a norm violation, which allows the agent to stay operational. Based on answer set programming (ASP), emergency plans are generated that exclude or minimize cost of norm violations by future actions in a horizon of interest. By combining and developing optimization techniques, efficient policy fixing under real-time constraints can be achieved.

Temporal Equilibrium Logic (TEL) extends Answer Set Programming (ASP) with linear-time temporal operators (LTL), enabling reasoning about dynamic systems. However, TEL enforces strong minimization criteria that may preclude intuitive models. Liveness formulas, for instance, tend to fail to have infinite equilibrium models, as TEL minimization postpones satisfaction forever. We address this limitation by introducing eager temporal operators (eager Until, eager Release, etc.), and present non-disjunctive temporal programs (NDTP) as a framework for modeling dependencies, inertia, and non-determinism. The fragment of tight temporal programs (TTP), which can be recognized efficiently based on automata techniques for loop detections, guarantees polynomial encodability into LTL. Practical examples, such as request-grant protocols and user permissions in distributed systems, illustrate the applicability of our approach.

We present the solver asp-fzn for Constraint Answer Set Programming (CASP), which extends ASP with linear constraints. Our approach is based on translating CASP programs into the solver-independent FlatZinc language that supports several Constraint Programming and Integer Programming backend solvers. Our solver supports a rich language of linear constraints, including some common global constraints. As for evaluation, we show that asp-fzn is competitive with state-of-the-art ASP solvers on benchmarks taken from past ASP competitions. Furthermore, we evaluate it on several CASP problems from the literature and compare its performance with clingcon, which is a prominent CASP solver that supports most of the asp-fzn language. The performance of asp-fzn is very promising as it is already competitive on plain ASP and even outperforms clingcon on some CASP benchmarks.

Non-monotonic logic programming is the basis for a declarative problem solving paradigm known as answer set programming (ASP). Departing from the seminal definition by Gelfond and Lifschitz in 1988 for simple normal logic programs, various answer set semantics have been proposed for extensions. We consider two important questions: (1) Should the minimal model property, constraint monotonicity and foundedness as defined in the literature be mandatory conditions for an answer set semantics in general? (2) If not, what other properties could be considered as general principles for answer set semantics? We address the two questions. First, it seems that the three aforementioned conditions may sometimes be too strong, and we illustrate with examples that enforcing them may exclude expected answer sets. Second, we evolve the Gelfond answer set (GAS) principles for answer set construction by refining the Gelfond's rationality principle to well-supportedness, minimality w.r.t. negation by default and minimality w.r.t. epistemic negation. The principle of well-supportedness guarantees that every answer set is constructible from if-then rules obeying a level mapping and is thus free of circular justification, while the two minimality principles ensure that the formalism minimizes knowledge both at the level of answer sets and of world views. Third, to embody the refined GAS principles, we extend the notion of well-supportedness substantially to answer sets and world views, respectively. Fourth, we define new answer set semantics in terms of the refined GAS principles. Fifth, we use the refined GAS principles as an alternative baseline to intuitively assess the existing answer set semantics. Finally, we analyze the computational complexity.

Declarative languages are widely used for reasoning about actions and planning, with semantics and extensions to cater for different needs. We introduce the rule-based action language deon-B, which builds on an extension of the well-founded semantics by incorporating deontic operators, allowing one to reason about obligations, prohibitions, and permissions in a non-monotonic setting. To further enhance its applicability in dynamic environments, we integrate a mechanism for action choice inspired by answer set-based action languages, enabling reasoning about transitions between states and for expressing the evolution of normative goals. Our approach provides a computationally efficient approximation of answer-set-based deontic reasoning while preserving key expressiveness. We evaluate the proposed formalism against well-known deontic challenges such as contrary-to-duty obligations and the distinction between prima facie and actual obligations, demonstrating its ability to capture and resolve normative conflicts within dynamic domains.

Answer Set Programming (ASP) is a popular nonmonotonic formalism used for common-sense reasoning and problem-solving based on stable model semantics. Equilibrium logic is a generalisation of ASP for arbitrary propositional theories and thus provides a logical characterisation of the nonmonotonic stable model semantics. In difference to classical logic, which can be defined via proof or model theory, nonmonotonic reasoning formalisms are defined via their models exclusively. Equilibrium logic is no exception here, as it has no proper proof-theoretic axiomatisation. Besides this being a theoretical imbalance, it also has consequences regarding notions of justification and explainability. In this work, we fill this gap by providing a sequent calculus for answer set entailment. Our calculus builds upon ideas from existing calculi for other nonmonotonic formalisms and utilises calculi for the logic of here and there, which is the underlying base logic of equilibrium logic. We show that the calculus is sound and complete and discuss pitfalls as well as alternative axiomatisations. Finally, we address how our approach can be of use for explainability in ASP.

Explanation plays an important role in the decisions of both symbolic and neural network-based AI systems. Logic programs under answer set semantics (ASP) have been a typical declarative reasoning and problem-solving paradigm that has extensive applications in various AI domains. In this paper, we consider the issue of explanation for logic programs with abstract constraint atoms (c-atoms) under SPT-answer set semantics. Such c-atoms are general enough to capture complex constructors of logic programs, including aggregates, and the SPT-answer sets exclude circular justifications that other semantics have. We propose a minimal reduct for logic programs with c-atoms that yields a new semantic characterization of SPT-answer sets, and then introduce an extension of resolution for clauses with c-atoms. As we show, every atom in an SPT-answer set enjoys an extended resolution proof from the minimal reduct of its logic program. Finally, we present minimal sufficient subsets of logic programs (witnesses) to structure such an extended resolution proof for an atom in an SPT-answer set. Our results contribute to the justification of answer sets and provide a basis for explainability of ASP-based applications.