The topic of comprehensibility of machine-learned theories has recently drawn increasing attention. inductive logic programming uses logicprogramming to derive logic theories from small data based on abduction and in...
详细信息
The topic of comprehensibility of machine-learned theories has recently drawn increasing attention. inductive logic programming uses logicprogramming to derive logic theories from small data based on abduction and induction techniques. Learned theories are represented in the form of rules as declarative descriptions of obtained knowledge. In earlier work, the authors provided the first evidence of a measurable increase in human comprehension based on machine-learned logic rules for simple classification tasks. In a later study, it was found that the presentation of machine-learned explanations to humans can produce both beneficial and harmful effects in the context of game learning. We continue our investigation of comprehensibility by examining the effects of the ordering of concept presentations on human comprehension. In this work, we examine the explanatory effects of curriculum order and the presence of machine-learned explanations for sequential problem-solving. We show that (1) there exist tasks A and B such that learning A before learning B results in better comprehension for humans in comparison to learning B before learning A and (2) there exist tasks A and B such that the presence of explanations when learning A contributes to improved human comprehension when subsequently learning B. We propose a framework for the effects of sequential teaching on comprehension based on an existing definition of comprehensibility and provide evidence for support from data collected in human trials. Our empirical study involves curricula that teach novices the merge sort algorithm. Our results show that sequential teaching of concepts with increasing complexity (a) has a beneficial effect on human comprehension and (b) leads to human re-discovery of divide-and-conquer problem-solving strategies, and (c) allows adaptations of human problem-solving strategy with better performance when machine-learned explanations are also presented.
Petri nets are a class of models of computation used to compactly represent discrete event systems. Among many application domains, they have now become the most prominent formalism to express process models in Proces...
详细信息
ISBN:
(纸本)9783031742088;9783031742095
Petri nets are a class of models of computation used to compactly represent discrete event systems. Among many application domains, they have now become the most prominent formalism to express process models in Process Mining, thanks to their formal semantics that enables automated analysis techniques. In this context, model repair is the task of aligning a process model with actual executions of the process. Current solutions to model repair do not allow for embedding domain knowledge, providing guarantees of rigor, and enforcing structural requirements at the same time. In this paper, we fill this gap by proposing an approach based on the inductive logic programming system ILASP. We then implement our approach and perform an experimental evaluation, showing both its expressiveness and feasibility.
Given a set of candidate Datalog rules, the Datalog synthesis-as-rule-selection problem chooses a subset of these rules that satisfies a specification (such as an input-output example). Building off prior work using c...
详细信息
Given a set of candidate Datalog rules, the Datalog synthesis-as-rule-selection problem chooses a subset of these rules that satisfies a specification (such as an input-output example). Building off prior work using counterexample-guided inductive synthesis, we present a progression of three solver-based approaches for solving Datalog synthesis-as-rule-selection problems. Two of our approaches offer some advantages over existing approaches, and can be used more generally to solve arbitrary SMT formulas containing Datalog predicates;the third-an encoding into standard, off-the-shelf answer set programming (ASP)-leads to significant speedups (similar to 9x geomean) over the state of the art while synthesizing higher quality programs. Our progression of solutions explores the space of interactions between SAT/SMT and Datalog, identifying ASP as a promising tool for working with and reasoning about Datalog. Along the way, we identify Datalog programs as monotonic SMT theories, which enjoy particularly efficient interactions in SMT;our plugins for popular SMT solvers make it easy to load an arbitrary Datalog program into the SMT solver as a custom monotonic theory. Finally, we evaluate our approaches using multiple underlying solvers to provide a more thorough and nuanced comparison against the current state of the art.
The goal of inductive logic programming is to learn a logic program that models the examples provided as input. The search space of the possible programs is constrained by a language bias, which defines the atoms and ...
详细信息
The goal of inductive logic programming is to learn a logic program that models the examples provided as input. The search space of the possible programs is constrained by a language bias, which defines the atoms and literals allowed in rules. Answer set programming is a powerful formalism to represent complex combinatorial domains, also thanks to syntactic constructs such as aggregates. However, learning answer set programs from data is challenging, and often existing tools do not support the specification of aggregates in the language bias. In this paper, we introduce GENTIANS, a tool based on a genetic algorithm to learn answer set programs possibly with aggregates, arithmetic, and comparison operators, from examples. Empirical results, also against an existing solver, show that GENTIANS is able to provide accurate solutions even when the search space contains millions of clauses. Additionally, experiments on noisy datasets show the effectiveness of our approach.
There have been attempts to connect machine learning and symbolic reasoning, providing interfaces between them. This work focuses on our original approach to integrate machine learning and symbolic reasoning, in the c...
详细信息
ISBN:
(纸本)9789819722990;9789819723003
There have been attempts to connect machine learning and symbolic reasoning, providing interfaces between them. This work focuses on our original approach to integrate machine learning and symbolic reasoning, in the context of algebraic approaches to logicprogramming. We here realize logical reasoning using algebraic methods, in which algebraic data structures such as matrices and tensors are used to represent logical formulas. These reasoning methods are robust against noise, while allowing for high parallelism and scalable computation. Algebraic logicprogramming has been applied to fixponit computation, abduction, answer set programming and inductive logic programming.
Explaining concepts by contrasting examples is an efficient and convenient way of giving insights into the reasons behind a classification decision. This is of particular interest in decision-critical domains, such as...
详细信息
ISBN:
(数字)9783031556302
ISBN:
(纸本)9783031556296;9783031556302
Explaining concepts by contrasting examples is an efficient and convenient way of giving insights into the reasons behind a classification decision. This is of particular interest in decision-critical domains, such as medical diagnostics. One particular challenging use case is to distinguish facial expressions of pain and other states, such as disgust, due to high similarity of manifestation. In this paper, we present an approach for generating contrastive explanations to explain facial expressions of pain and disgust shown in video sequences. We implement and compare two approaches for contrastive explanation generation. The first approach explains a specific pain instance in contrast to the most similar disgust instance(s) based on the occurrence of facial expressions (attributes). The second approach takes into account which temporal relations hold between intervals of facial expressions within a sequence (relations). The input to our explanation generation approach is the output of an interpretable rule-based classifier for pain and disgust. We utilize two different similarity metrics to determine near misses and far misses as contrasting instances. Our results show that near miss explanations are shorter than far miss explanations, independent from the applied similarity metric. The outcome of our evaluation indicates that pain and disgust can be distinguished with the help of temporal relations. We currently plan experiments to evaluate how the explanations help in teaching concepts and how they could be enhanced by further modalities and interaction.
While remarkable recent developments in deep neural networks have significantly contributed to advancing the state-of-the-art in Computer Vision (CV), several studies have also shown their limitations and defects. In ...
详细信息
ISBN:
(纸本)9783031711664;9783031711671
While remarkable recent developments in deep neural networks have significantly contributed to advancing the state-of-the-art in Computer Vision (CV), several studies have also shown their limitations and defects. In particular, CV models often make systematic errors on important subsets of data called slices, which are groups of data sharing a set of attributes. The slice discovery problem involves detecting semantically meaningful slices on which the model performs poorly, called rare slices. We propose a modular Neurosymbolic AI approach whose distinct advantage is the extraction of human-readable logical rules that describe rare slices, and thus enhances explainability of CV models. To this end, we present a methodology to induce rare slice occurrences in a model. Experiments on datasets from our data generator leveraging on Super-CLEVR show that the approach can correctly identify rare slices and produce logical rules describing them. The rules can be fruitfully used to generate new training data to mend model behavior or may be integrated into the model to enhance its inference capabilities. (The code for reproducing our experiments is available as an online repository: https://***/kbs/nesy- ai/ilp4sd).
The analysis of structure in complex datasets has become essential to solving difficult Machine Learning problems. Relational aspects of data, capturing relationships between objects, play a crucial role in understand...
详细信息
ISBN:
(纸本)9798350359329;9798350359312
The analysis of structure in complex datasets has become essential to solving difficult Machine Learning problems. Relational aspects of data, capturing relationships between objects, play a crucial role in understanding the underlying data structure. While traditional graph algorithms have been widely used for binary relations, recent evidence suggests that hypergraphs can provide a more effective approach for modeling complex, non-binary relations. Hypergraph Neural Networks (HGNN) have been shown to offer a small improvement in performance when compared to Graph Neural Networks (GNN). In this paper, a new approach is proposed for inserting relational domain knowledge into HGNNs using a logic clause expressing non-binary relations. We evaluate the performance of this new hypergraph model, called Bottom-clause HGNN (BHGNN), in comparison with well-known approaches. Results show that BHGNN can achieve statistically significant improvement of performance, based on the Wilcoxon signed-ranks test, in comparison with HGNN and GNNs.
Learning from interpretation transition (LFIT) automatically constructs a model of the dynamics of a system from the observation of its state transitions. So far the systems that LFIT handled were mainly restricted to...
详细信息
Learning from interpretation transition (LFIT) automatically constructs a model of the dynamics of a system from the observation of its state transitions. So far the systems that LFIT handled were mainly restricted to synchronous deterministic dynamics. However, other dynamics exist in the field of logical modeling, in particular the asynchronous semantics which is widely used to model biological systems. In this paper, we propose a modeling of discrete memory-less multi-valued dynamic systems as logic programs in which a rule represents what can occur rather than what will occur. This modeling allows us to represent non-determinism and to propose an extension of LFIT to learn regardless of the update schemes, allowing to capture a large range of semantics. We also propose a second algorithm which is able to learn a whole system dynamics, including its semantics, in the form of a single propositional logic program with constraints. We show through theoretical results the correctness of our approaches. Practical evaluation is performed on benchmarks from biological literature.
We propose the identification of feedback mechanisms in biological systems by learning logical rules in R. Thomas' Kinetic logic (Thomas and D'Ari in Biological feedback. CRC Press, 1990). The principal advant...
详细信息
We propose the identification of feedback mechanisms in biological systems by learning logical rules in R. Thomas' Kinetic logic (Thomas and D'Ari in Biological feedback. CRC Press, 1990). The principal advantages claimed for Kinetic logic are that it captures an important class of regulatory networks at an appropriate level of precision, and that the representation is close to that used routinely by biologists, with a well-understood relationship to a differential description. In this paper we present a formalisation of Kinetic logic as a labelled transition system and provide a provably correct implementation in a modified form of the Event Calculus. The behaviour of a system is then a logical consequence of the core-axioms of a (modified) Event Calculus C, the axioms K implementing Kinetic logic and the axioms H describing the system. This formulation allows us to specify system identification in the manner adopted in inductive logic programming (ILP), namely, given C, K, system behaviour S and possibly some additional domain-knowledge B, find H s.t. B <^> C <^> K <^> H (sic) S. Identifying a suitable Kinetic logic hypothesis requires the simultaneous identification of definite clauses for: (a) logical definitions relating the occurrence of events to values of fluents;(b) delays in changes of the values of fluents arising from the occurrence of events;and possibly (c) exceptions to changes in fluent values, arising from asynchronous behaviour inherent to the system. We use a standard ILP engine for (a), and special-purpose abduction procedures for (b) and (c). We demonstrate this combination of induction and abduction on several canonical feedback patterns described by Thomas, and to identify the regulatory mechanism in two well-known biological problems (immuneresponse and phage-infection).
暂无评论