检索结果-内蒙古大学图书馆

4th International Workshop on Explainable and Transparent AI and Multi-Agent Systems (EXTRAAMAS)

作者： Beretta, Davide Monica, Stefania Bergenti, Federico Univ Parma Dipartimento Sci Matemat Fis & Informat I-43124 Parma Italy Univ Modena & Reggio Emilia Dipartimento Sci & Metodi Ingn I-42122 Reggio Emilia Italy

ISBN: (纸本)9783031155659;9783031155642

Deep learning has been increasingly successful in the last few years, but its inherent limitations have recently become more evident, especially with respect to explainability and interpretability. Neural-symbolic approaches to inductive logic programming have been recently proposed to synergistically combine the advantages of inductive logic programming in terms of explainability and interpretability with the characteristic capability of deep learning to treat noisy, erroneous, and non-logical data. This paper surveys and briefly compares four relevant neural-symbolic approaches to inductive logic programming that have been proposed in the last five years and that use templates as an effective basis to learn logic programs from data.

关键词： Neural-symbolic learning inductive logic programming Explainable artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Applied to Harmonic Functions in Music Composition 7th

Machine Learning Applied to Harmonic Functions in Music Comp...

引用

7th Brazilian Technology Symposium (BTSym) - Emerging Trends and Challenges in Technology

作者： Goncalves Junior, Clenio B. Petrucelli Homem, Murillo R. Fed Inst Educ Sci & Technol Sao Paulo Sao Paulo Brazil Univ Fed Sao Carlos Dept Comp Sci Sao Carlos Brazil

ISBN: (纸本)9783031085451;9783031085444

The knowledge representation process in Computer Music is an essential element for the development of systems. Methods have been applied to provide the computer with the ability to infer information from previously established experience and definitions. Alongside this process, with regard to the aspect of musical interaction, it is observed that in the generation of automatic musical compositions, the harmonic component is fundamental in the production of associations aimed at human interaction. However, there is a lack of studies that deal with the functions performed by chords. In this sense, inductive logic programming is a growing field of research that incorporates concepts of logic programming and Machine Learning. This work consists in the application of the technique in Machine Learning of inductive logic programming, performing the derivation of harmonic functions. Musical rules are represented by First-Order logic and a corpus-based on Functional Harmony theory is used as background knowledge. Through the experimental method, the induction of 3 functional rules was performed. The measures of precision, coverage, accuracy, and execution time indicate the feasibility of this approach according to the purpose of algorithmic musical composition.

关键词： Machine learning Harmonic functions inductive logic programming Music composition

来源：评论

学校读者我要写书评

暂无评论

Online Learning of logic Based Neural Network Structures 1

引用

30th International Conference on inductive logic programming (ILP) held as part of the 1st International Joint Conference on Learning and Reasoning (IJCLR)

作者： Guimaraes, Victor Costa, Vitor Santos Univ Porto CRACS Porto Portugal Univ Porto DCC FCUP Porto Portugal

ISBN: (数字)9783030974541

ISBN: (纸本)9783030974541;9783030974534

In this paper, we present two online structure learning algorithms for NeuralLog, NeuralLog+OSLR and NeuralLog+OMIL. NeuralLog is a system that compiles first-order logic programs into neural networks. Both learning algorithms are based on Online Structure Learner by Revision (OSLR). NeuralLog+OSLR is a port of OSLR to use NeuralLog as inference engine;while NeuralLog+OMIL uses the underlying mechanism from OSLR, but with a revision operator based on Meta-Interpretive Learning. We compared both systems with OSLR and RDN-Boost on link prediction in three different datasets: Cora, UMLS and UWCSE. Our experiments showed that NeuralLog+OMIL outperforms both the compared systems on three of the four target relations from the Cora dataset and in the UMLS dataset, while both NeuralLog+OSLR and NeuralLog+OMIL outperform OSLR and RDNBoost on the UWCSE, assuming a good initial theory is provided.

关键词： Online Learning inductive logic programming Meta-Interpretive Learning Neural Network Neural-Symbolic Learning and Reasoning Theory Revision from Examples

来源：评论

学校读者我要写书评

暂无评论

Learning to Rank the Distinctiveness of Behaviour in Serial Offending 16th

Learning to Rank the Distinctiveness of Behaviour in Serial ...

引用

16th International Conference on logic programming and Non-Monotonic Reasoning (LPNMR)

作者： Law, Mark Sautory, Theophile Mitchener, Ludovico Davies, Kari Tonkin, Matthew Woodhams, Jessica Alrajeh, Dalal ILASP Ltd Grantham England Univ Calif Berkeley Berkeley CA 94720 USA Imperial Coll London London England Univ Bournemouth Bournemouth Dorset England Univ Leicester Leicester Leics England Univ Birmingham Birmingham W Midlands England

ISBN: (纸本)9783031157073;9783031157066

Comparative Case Analysis is an analytical process used to detect serial offending. It focuses on identifying distinctive behaviour that an offender displays consistently when committing their crimes. In practice, crime analysts consider the context in which each behaviour occurs to determine its distinctiveness, which subsequently impacts on their determination of whether crimes are committed by the same person or not. Existing algorithms do not currently consider context in this way when generating linkage predictions. This paper presents the first learning-based approach aimed at identifying contexts within which behaviour may be considered more distinctive. We show how this problem can be modelled as that of learning preferences (in answer set programming) from examples of ordered pairs of contexts in which a behaviour was observed. In this setting, a context is preferred to another context if the behaviour is rarer in the first context. We make novel use of odds ratios to determine which examples are used for learning. Our approach has been applied to a real dataset of serious sexual offences provided by the UK National Crime Agency. The approach provides (i) a systematic methodology for selecting examples from which to learn preferences;(ii) novel insights for practitioners into the contexts under which an exhibited behaviour is more rare.

关键词： Crime Linkage Answer Set programming inductive logic programming

来源：评论

学校读者我要写书评

暂无评论

Learning programs by learning from failures

引用

MACHINE LEARNING 2021年第4期110卷 801-856页

作者： Cropper, Andrew Morel, Rolf Univ Oxford Oxford England

We describe an inductive logic programming (ILP) approach called learning from failures. In this approach, an ILP system (the learner) decomposes the learning problem into three separate stages: generate, test, and constrain. In the generate stage, the learner generates a hypothesis (a logic program) that satisfies a set of hypothesis constraints (constraints on the syntactic form of hypotheses). In the test stage, the learner tests the hypothesis against training examples. A hypothesis fails when it does not entail all the positive examples or entails a negative example. If a hypothesis fails, then, in the constrain stage, the learner learns constraints from the failed hypothesis to prune the hypothesis space, i.e. to constrain subsequent hypothesis generation. For instance, if a hypothesis is too general (entails a negative example), the constraints prune generalisations of the hypothesis. If a hypothesis is too specific (does not entail all the positive examples), the constraints prune specialisations of the hypothesis. This loop repeats until either (i) the learner finds a hypothesis that entails all the positive and none of the negative examples, or (ii) there are no more hypotheses to test. We introduce Popper, an ILP system that implements this approach by combining answer set programming and Prolog. Popper supports infinite problem domains, reasoning about lists and numbers, learning textually minimal programs, and learning recursive programs. Our experimental results on three domains (toy game problems, robot strategies, and list transformations) show that (i) constraints drastically improve learning performance, and (ii) Popper can outperform existing ILP systems, both in terms of predictive accuracies and learning times.

关键词： inductive logic programming Program synthesis Program induction

来源：评论

学校读者我要写书评

暂无评论

Rule Learning over Knowledge Graphs with Genetic logic programming 38

Rule Learning over Knowledge Graphs with Genetic Logic Progr...

引用

38th IEEE International Conference on Data Engineering (ICDE)

作者： Wu, Lianlong Sallinger, Emanuel Sherkhonov, Evgeny Vahdati, Sahar Gottlob, Georg Univ Oxford Dept Comp Sci Oxford England TU Wien Inst Log & Computat Vienna Austria

ISBN: (数字)9781665408837

ISBN: (纸本)9781665408837

Declarative rules such as Prolog and Datalog rules are common formalisms to express expert knowledge and facts. They play an important role in Knowledge Graph (KG) construction and completion. Such rules not only encode the expert background knowledge and the relational patterns among the data, but also infer new knowledge and insights from them. Formalizing rules is often a laborious manual process, while learning them from data automatically can ease this process. Within the rule hypothesis space, current approaches resort to exhaustive search with a number of heuristics and syntactic restrictions on the rule language, which impacts the efficiency and quality of the outcome rules. In this paper, we extend the rule hypothesis space from usual path rules to general Datalog rule space by proposing a novel Genetic logic programming algorithm named Evoda. It is an iterative process to learn high-quality rules over large scale KG for a matter of seconds. We have performed experiments over multiple real-world KGs and various evaluation metrics to show its mining capabilities for higher quality rules and more precise predictions. Additionally, we have applied it on the KG completion tasks to illustrate its competitiveness with several state-of-the-art embedding or neural-based models. The experiments demonstrate the feasibility, effectiveness and efficiency of the Evoda algorithm.

关键词： Rule Mining Knowledge Graph inductive logic programming Genetic programming

来源：评论

学校读者我要写书评

暂无评论

inductive learning of answer set programs for autonomous surgical task planning Application to a training task for surgeons

引用

MACHINE LEARNING 2021年第7期110卷 1739-1763页

作者： Meli, Daniele Sridharan, Mohan Fiorini, Paolo Univ Verona Dept Comp Sci Str Grazie 15 I-37135 Verona Italy Univ Birmingham Sch Comp Sci Birmingham B15 2TT W Midlands England

The quality of robot-assisted surgery can be improved and the use of hospital resources can be optimized by enhancing autonomy and reliability in the robot's operation. logic programming is a good choice for task planning in robot-assisted surgery because it supports reliable reasoning with domain knowledge and increases transparency in the decision making. However, prior knowledge of the task and the domain is typically incomplete, and it often needs to be refined from executions of the surgical task(s) under consideration to avoid sub-optimal performance. In this paper, we investigate the applicability of inductive logic programming for learning previously unknown axioms governing domain dynamics. We do so under answer set semantics for a benchmark surgical training task, the ring transfer. We extend our previous work on learning the immediate preconditions of actions and constraints, to also learn axioms encoding arbitrary temporal delays between atoms that are effects of actions under the event calculus formalism. We propose a systematic approach for learning the specifications of a generic robotic task under the answer set semantics, allowing easy knowledge refinement with iterative learning. In the context of 1000 simulated scenarios, we demonstrate the significant improvement in performance obtained with the learned axioms compared with the hand-written ones;specifically, the learned axioms address some critical issues related to the plan computation time, which is promising for reliable real-time performance during surgery.

关键词： inductive logic programming Surgical robotics Answer set programming

来源：评论

学校读者我要写书评

暂无评论

Top program construction and reduction for polynomial time Meta-Interpretive learning

引用

MACHINE LEARNING 2021年第4期110卷 755-778页

作者： Patsantzis, S. Muggleton, S. H. Imperial Coll London London England

Meta-Interpretive Learners, like most ILP systems, learn by searching for a correct hypothesis in the hypothesis space, the powerset of all constructible clauses. We show how this exponentially-growing search can be replaced by the construction of a Top program: the set of clauses in all correct hypotheses that is itself a correct hypothesis. We give an algorithm for Top program construction and show that it constructs a correct Top program in polynomial time and from a finite number of examples. We implement our algorithm in Prolog as the basis of a new MIL system, Louise, that constructs a Top program and then reduces it by removing redundant clauses. We compare Louise to the state-of-the-art search-based MIL system Metagol in experiments on grid world navigation, graph connectedness and grammar learning datasets and find that Louise improves on Metagol's predictive accuracy when the hypothesis space and the target theory are both large, or when the hypothesis space does not include a correct hypothesis because of "classification noise" in the form of mislabelled examples. When the hypothesis space or the target theory are small, Louise and Metagol perform equally well.

关键词： inductive logic programming Meta interpretive learning Machine learning Top program construction

来源：评论

学校读者我要写书评

暂无评论

Beneficial and harmful explanatory machine learning

引用

MACHINE LEARNING 2021年第4期110卷 695-721页

作者： Ai, Lun Muggleton, Stephen H. Hocquette, Celine Gromowski, Mark Schmid, Ute Imperial Coll London Dept Comp London England Univ Bamberg Cognit Syst Grp Bamberg Germany

Given the recent successes of Deep Learning in AI there has been increased interest in the role and need for explanations in machine learned theories. A distinct notion in this context is that of Michie's definition of ultra-strong machine learning (USML). USML is demonstrated by a measurable increase in human performance of a task following provision to the human of a symbolic machine learned theory for task performance. A recent paper demonstrates the beneficial effect of a machine learned logic theory for a classification task, yet no existing work to our knowledge has examined the potential harmfulness of machine's involvement for human comprehension during learning. This paper investigates the explanatory effects of a machine learned theory in the context of simple two person games and proposes a framework for identifying the harmfulness of machine explanations based on the Cognitive Science literature. The approach involves a cognitive window consisting of two quantifiable bounds and it is supported by empirical evidence collected from human trials. Our quantitative and qualitative results indicate that human learning aided by a symbolic machine learned theory which satisfies a cognitive window has achieved significantly higher performance than human self learning. Results also demonstrate that human learning aided by a symbolic machine learned theory that fails to satisfy this window leads to significantly worse performance than unaided human learning.

关键词： inductive logic programming Comprehensibility Ultra-strong machine learning Explainable AI

来源：评论

学校读者我要写书评

暂无评论

aILP: thinking visual scenes as differentiable logic programs

引用

MACHINE LEARNING 2023年第5期112卷 1465-1497页

作者： Shindo, Hikaru Pfanschilling, Viktor Dhami, Devendra Singh Kersting, Kristian Tech Univ Darmstadt Darmstadt Germany Hessian Ctr AI Hessian AI Darmstadt Germany Tech Univ Darmstadt Ctr Cognit Sci Darmstadt Germany

Deep neural learning has shown remarkable performance at learning representations for visual object categorization. However, deep neural networks such as CNNs do not explicitly encode objects and relations among them. This limits their success on tasks that require a deep logical understanding of visual scenes, such as Kandinsky patterns and Bongard problems. To overcome these limitations, we introduce aILP, a novel differentiable inductive logic programming framework that learns to represent scenes as logic programs & mdash;intuitively, logical atoms correspond to objects, attributes, and relations, and clauses encode high-level scene information. aILP has an end-to-end reasoning architecture from visual inputs. Using it, aILP performs differentiable inductive logic programming on complex visual scenes, i.e., the logical rules are learned by gradient descent. Our extensive experiments on Kandinsky patterns and CLEVR-Hans benchmarks demonstrate the accuracy and efficiency of aILP in learning complex visual-logical concepts.

关键词： Differentiable reasoning inductive logic programming Object-centric learning Neuro-symbolic AI

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：