Publications
280 results found
, 2024, Inductive Logic Programming - 31st International Conference, ILP 2022, Windsor Great Park, UK, September 28-30, 2022, Proceedings, Publisher: Springer
Ai L, Langer J, Muggleton SH, et al., 2023, Explanatory machine learning for sequential human teaching., Mach. Learn., Vol: 112, Pages: 3591-3632
Bundy A, Chater N, Muggleton S, 2023, Introduction to 'Cognitive artificial intelligence'., Philos Trans A Math Phys Eng Sci, Vol: 381
Ai L, Langer J, Muggleton SH, et al., 2023, Explanatory machine learning for sequential human teaching, MACHINE LEARNING, ISSN: 0885-6125
Patsantzis S, Muggleton SH, 2022, Meta-interpretive learning as metarule specialisation, MACHINE LEARNING, Vol: 111, Pages: 3703-3731, ISSN: 0885-6125
- Author Web Link
- Cite
- Citations: 2
Patsantzis S, Muggleton SH, 2022, Meta-interpretive learning as metarule specialisation (Apr, 10.1007/s10994-022-06156-1, 2022), MACHINE LEARNING, Vol: 111, Pages: 3061-3061, ISSN: 0885-6125
Cropper A, Dumancic S, Evans R, et al., 2022, Inductive logic programming at 30, Publisher: SPRINGER, Pages: 147-172, ISSN: 0885-6125
- Author Web Link
- Cite
- Citations: 10
Barroso-Bergada D, Tamaddoni-Nezhad A, Muggleton SH, et al., 2022, Machine Learning of Microbial Interactions Using Abductive ILP and Hypothesis Frequency/Compression Estimation, 30th International Conference on Inductive Logic Programming (ILP) held as part of the 1st International Joint Conference on Learning and Reasoning (IJCLR), Publisher: SPRINGER INTERNATIONAL PUBLISHING AG, Pages: 26-40, ISSN: 0302-9743
- Author Web Link
- Cite
- Citations: 2
Varghese D, Bauer R, Baxter-Beard D, et al., 2022, Human-Like Rule Learning from Images Using One-Shot Hypothesis Derivation, 30th International Conference on Inductive Logic Programming (ILP) held as part of the 1st International Joint Conference on Learning and Reasoning (IJCLR), Publisher: SPRINGER INTERNATIONAL PUBLISHING AG, Pages: 234-250, ISSN: 0302-9743
Tamaddoni-Nezhad A, Bohan D, Milani GA, et al., 2021, Human-machine scientific discovery, Human-Like Machine Intelligence, Pages: 297-315, ISBN: 9780198862536
- Cite
- Citations: 2
Muggleton S, Dai WZ, 2021, Human-like computer vision, Human-Like Machine Intelligence, Pages: 199-217, ISBN: 9780198862536
Dai W-Z, Muggleton S, 2021, Abductive knowledge induction from raw data, Thirtieth International Joint Conference on Artificial Intelligence {IJCAI-21}, Publisher: International Joint Conferences on Artificial Intelligence Organization, Pages: 1845-1851
For many reasoning-heavy tasks with raw inputs, it is challenging to design an appropriate end-to-end pipeline to formulate the problem-solving process. Some modern AI systems, e.g., Neuro-Symbolic Learning, divide the pipeline into sub-symbolic perception and symbolic reasoning, trying to utilise data-driven machine learning and knowledge-driven problem-solving simultaneously. However, these systems suffer from the exponential computational complexity caused by the interface between the two components, where the sub-symbolic learning model lacks direct supervision, and the symbolic model lacks accurate input facts. Hence, they usually focus on learning the sub-symbolic model with a complete symbolic knowledge base while avoiding a crucial problem: where does the knowledge come from? In this paper, we present Abductive Meta-Interpretive Learning (MetaAbd) that unites abduction and induction to learn neural networks and logic theories jointly from raw data. Experimental results demonstrate that MetaAbd not only outperforms the compared systems in predictive accuracy and data efficiency but also induces logic programs that can be re-used as background knowledge in subsequent learning tasks. To the best of our knowledge, MetaAbd is the first system that can jointly learn neural networks from scratch and induce recursive first-order logic theories with predicate invention.
Muggleton S, Chater N, 2021, Human-Like Machine Intelligence, ISBN: 9780198862536
In recent years there has been increasing excitement concerning the potential of Artificial Intelligence to transform human society. This book addresses the leading edge of research in this area. The research described aims to address present incompatibilities of Human and Machine reasoning and learning approaches. According to the influential US funding agency DARPA (originator of the Internet and Self-Driving Cars) this new area represents the Third Wave of Artificial Intelligence (3AI, 2020s-2030s), and is being actively investigated in the US, Europe and China. The EPSRC's UK network on Human-Like Computing (HLC) was one of the first internationally to initiate and support research specifically in this area. Starting activities in 2018, the network represents around sixty leading UK groups Artificial Intelligence and Cognitive Scientists involved in the development of the inter-disciplinary area of HLC. The research of network groups aims to address key unsolved problems at the interface between Psychology and Computer Science. The chapters of this book have been authored by a mixture of these UK and other international specialists based on recent workshops and discussions at the Machine Intelligence 20 and 21 workshops (2016,2019) and the Third Wave Artificial Intelligence workshop (2019). Some of the key questions addressed by the Human-Like Computing programme include how AI systems might 1) explain their decisions effectively, 2) interact with human beings in natural language, 3) learn from small numbers of examples and 4) learn with minimal supervision. Solving such fundamental problems involves new foundational research in both the Psychology of perception and interaction as well as the development of novel algorithmic approaches in Artificial Intelligence.
Dai W-Z, Hallett L, Muggleton SH, et al., 2021, Automated Biodesign Engineering by Abductive Meta-Interpretive Learning
The application of Artificial Intelligence (AI) to synthetic biology willprovide the foundation for the creation of a high throughput automated platformfor genetic design, in which a learning machine is used to iteratively optimisethe system through a design-build-test-learn (DBTL) cycle. However, mainstreammachine learning techniques represented by deep learning lacks the capabilityto represent relational knowledge and requires prodigious amounts of annotatedtraining data. These drawbacks strongly restrict AI's role in synthetic biologyin which experimentation is inherently resource and time intensive. In thiswork, we propose an automated biodesign engineering framework empowered byAbductive Meta-Interpretive Learning ($Meta_{Abd}$), a novel machine learningapproach that combines symbolic and sub-symbolic machine learning, to furtherenhance the DBTL cycle by enabling the learning machine to 1) exploit domainknowledge and learn human-interpretable models that are expressed by formallanguages such as first-order logic; 2) simultaneously optimise the structureand parameters of the models to make accurate numerical predictions; 3) reducethe cost of experiments and effort on data annotation by actively generatinghypotheses and examples. To verify the effectiveness of $Meta_{Abd}$, we havemodelled a synthetic dataset for the production of proteins from a three geneoperon in a microbial host, which represents a common synthetic biologyproblem.
Patsantzis S, Muggleton SH, 2021, Top program construction and reduction for polynomial time Meta-Interpretive learning, MACHINE LEARNING, Vol: 110, Pages: 755-778, ISSN: 0885-6125
- Author Web Link
- Cite
- Citations: 3
Ai L, Muggleton SH, Hocquette C, et al., 2021, Beneficial and harmful explanatory machine learning, MACHINE LEARNING, Vol: 110, Pages: 695-721, ISSN: 0885-6125
- Author Web Link
- Cite
- Citations: 14
Cai LW, Dai WZ, Huang YX, et al., 2021, Abductive Learning with Ground Knowledge Base, Pages: 1815-1821, ISSN: 1045-0823
Abductive Learning is a framework that combines machine learning with first-order logical reasoning. It allows machine learning models to exploit complex symbolic domain knowledge represented by first-order logic rules. However, it is challenging to obtain or express the ground-truth domain knowledge explicitly as first-order logic rules in many applications. The only accessible knowledge base is implicitly represented by groundings, i.e., propositions or atomic formulas without variables. This paper proposes Grounded Abductive Learning (GABL) to enhance machine learning models with abductive reasoning in a ground domain knowledge base, which offers inexact supervision through a set of logic propositions. We apply GABL on two weakly supervised learning problems and found that the model's initial accuracy plays a crucial role in learning. The results on a real-world OCR task show that GABL can significantly reduce the effort of data labeling than the compared methods.
Huang YX, Dai WZ, Cai LW, et al., 2021, Fast Abductive Learning by Similarity-based Consistency Optimization, Pages: 26574-26584, ISSN: 1049-5258
To utilize the raw inputs and symbolic knowledge simultaneously, some recent neuro-symbolic learning methods use abduction, i.e., abductive reasoning, to integrate sub-symbolic perception and logical inference. While the perception model, e.g., a neural network, outputs some facts that are inconsistent with the symbolic background knowledge base, abduction can help revise the incorrect perceived facts by minimizing the inconsistency between them and the background knowledge. However, to enable effective abduction, previous approaches need an initialized perception model that discriminates the input raw instances. This limits the application of these methods, as the discrimination ability is usually acquired from a thorough pre-training when the raw inputs are difficult to classify. In this paper, we propose a novel abduction strategy, which leverages the similarity between samples, rather than the output information by the perceptual neural network, to guide the search in abduction. Based on this principle, we further present ABductive Learning with Similarity (ABLSim) and apply it to some difficult neuro-symbolic learning tasks. Experiments show that the efficiency of ABLSim is significantly higher than the state-of-the-art neuro-symbolic methods, allowing it to achieve better performance with less labeled data and weaker domain knowledge.
Patsantzis S, Muggleton SH, 2021, Top program construction and reduction for polynomial time Meta-Interpretive learning., Mach. Learn., Vol: 110, Pages: 755-778
Cropper A, Dumancic S, Evans R, et al., 2021, Inductive logic programming at 30., CoRR, Vol: abs/2102.10556
Ai L, Muggleton SH, Hocquette C, et al., 2021, Beneficial and harmful explanatory machine learning., Mach. Learn., Vol: 110, Pages: 695-721
Cropper A, Morel R, Muggleton S, 2020, Learning higher-order logic programs, Machine Learning, Vol: 109, Pages: 1289-1322, ISSN: 0885-6125
A key feature of inductive logic programming is its ability to learn first-order programs, which are intrinsically more expressive than propositional programs. In this paper, we introduce techniques to learn higher-order programs. Specifically, we extend meta-interpretive learning (MIL) to support learning higher-order programs by allowing for higher-order definitions to be used as background knowledge. Our theoretical results show that learning higher-order programs, rather than first-order programs, can reduce the textual complexity required to express programs, which in turn reduces the size of the hypothesis space and sample complexity. We implement our idea in two new MIL systems: the Prolog system Metagol ho and the ASP system HEXMIL ho. Both systems support learning higher-order programs and higher-order predicate invention, such as inventing functions for map/3 and conditions for filter/3. We conduct experiments on four domains (robot strategies, chess playing, list transformations, and string decryption) that compare learning first-order and higher-order programs. Our experimental results support our theoretical claims and show that, compared to learning first-order programs, learning higher-order programs can significantly improve predictive accuracies and reduce learning times.
Hocquette C, Muggleton SH, 2020, Complete Bottom-Up Predicate Invention in Meta-Interpretive Learning, 29th International Joint Conference on Artificial Intelligence, Publisher: IJCAI-INT JOINT CONF ARTIF INTELL, Pages: 2312-2318
- Author Web Link
- Cite
- Citations: 5
Cropper A, Dumancic S, Muggleton SH, 2020, Turning 30: New Ideas in Inductive Logic Programming., Publisher: ijcai.org, Pages: 4833-4839
Cropper A, Dumancic S, Muggleton SH, 2020, Turning 30: New Ideas in Inductive Logic Programming, PROCEEDINGS OF THE TWENTY-NINTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, Pages: 4833-4839
- Author Web Link
- Cite
- Citations: 24
Cropper A, Morel R, Muggleton SH, 2020, Learning Higher-Order Programs through Predicate Invention, 34th AAAI Conference on Artificial Intelligence / 32nd Innovative Applications of Artificial Intelligence Conference / 10th AAAI Symposium on Educational Advances in Artificial Intelligence, Publisher: ASSOC ADVANCEMENT ARTIFICIAL INTELLIGENCE, Pages: 13655-13658, ISSN: 2159-5399
Ichise R, Muggleton S, Kozaki K, et al., 2019, Special Issue on Semantic Technology, New Generation Computing, Vol: 37, Pages: 359-360, ISSN: 0288-3635
Muggleton SH, Hocquette C, 2019, Machine discovery of comprehensible strategies for simple games using Meta-interpretive Learning, New Generation Computing, Vol: 37, Pages: 203-217, ISSN: 0288-3635
Recently, world-class human players have been outperformed in a number of complex two-person games (Go, Chess, Checkers) by Deep Reinforcement Learning systems. However, the data efficiency of the learning systems is unclear given that they appear to require far more training games to achieve such performance than any human player might experience in a lifetime. In addition, the resulting learned strategies are not in a form which can be communicated to human players. This contrasts to earlier research in Behavioural Cloning in which single-agent skills were machine learned in a symbolic language, facilitating their being taught to human beings. In this paper, we consider Machine Discovery of human-comprehensible strategies for simple two-person games (Noughts-and-Crosses and Hexapawn). One advantage of considering simple games is that there is a tractable approach to calculating minimax regret. We use these games to compare Cumulative Minimax Regret for variants of both standard and deep reinforcement learning against two variants of a new Meta-interpretive Learning system called MIGO. In our experiments, tested variants of both normal and deep reinforcement learning have consistently worse performance (higher cumulative minimax regret) than both variants of MIGO on Noughts-and-Crosses and Hexapawn. In addition, MIGO’s learned rules are relatively easy to comprehend, and are demonstrated to achieve significant transfer learning in both directions between Noughts-and-Crosses and Hexapawn.
Hocquette C, Muggleton SH, 2019, Can meta-interpretive learning outperform deep reinforcement learning of evaluable game strategies?, Publisher: arXiv
World-class human players have been outperformed in a number of complex two person games (Go, Chess, Checkers) by Deep Reinforcement Learning systems. However, owing to tractability considerations minimax regret of a learning system cannot be evaluated in such games. In this paper we consider simple games (Noughts-and-Crosses and Hexapawn) in which minimax regret can be efficiently evaluated. We use these games to compare Cumulative Minimax Regret for variants of both standard and deep reinforcement learning against two variants of a new Meta-Interpretive Learning system called MIGO. In our experiments all tested variants of both normal and deep reinforcement learning have worse performance (higher cumulative minimax regret) than both variants of MIGO on Noughts-and-Crosses and Hexapawn. Additionally, MIGO's learned rules are relatively easy to comprehend, and are demonstrated to achieve significant transfer learning in both directions between Noughts-and-Crosses and Hexapawn.
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.