Abductive Knowledge Induction from Raw Data
Open accessAbstract
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.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The authors demonstrate that their framework can learn arithmetic and sorting operations from digits represented as images.
“The authors demonstrate that their framework can learn arithmetic and sorting operations from digits represented as images.”
From Inductive Logic Programming At 30: A New Introduction (2022) · cites this paper as Dai & Muggleton, 2021
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Graph neural networks: A review of methods and applications
Zhou, J., Cui, G., Hu, S., Zhang, Z., Yang, C., Liu, Z., Wang, L., Li, C., & Sun, M. (2020) · AI Open · doi:10.1016/j.aiopen.2021.01.001 · Cited by 6,880 · see citing works
Inductive Logic Programming: Theory and methods
Muggleton, S., & de Raedt, L. (1994) · The Journal of Logic Programming · doi:10.1016/0743-1066(94)90035-3 · Cited by 1,537 · see citing works
Self-citationSatisfiability Modulo Theories
Barrett, C., & Tinelli, C. (2018) · Springer eBooks · doi:10.1007/978-3-319-10575-8_11 · Cited by 1,148 · see citing works
Kakas, A. C., KOWALSKI, R. A., & Toni, F. (1992) · Journal of Logic and Computation · doi:10.1093/logcom/2.6.719 · Cited by 833 · see citing works
Flach, P. A., & Kakas, A. C. (2000) · Applied logic series · doi:10.1007/978-94-017-0606-3 · Cited by 444 · see citing works
DeepProbLog: Neural Probabilistic Logic Programming
Manhaeve, R., Dumančić, S., Kimmig, A., Demeester, T., & De Raedt, L. (2018) · arXiv · doi:10.48550/arxiv.1805.10872 · Cited by 239 · see citing works
Lifschitz, V. (2019) · Springer eBooks · doi:10.1007/978-3-030-24658-7 · Cited by 233 · see citing works
Meta-interpretive learning of higher-order dyadic datalog: predicate invention revisited
Muggleton, S. H., Lin, D., & Tamaddoni-Nezhad, A. (2015) · Machine Learning · doi:10.1007/s10994-014-5471-y · Cited by 184 · see citing works
Self-citationLogic Tensor Networks for Semantic Image Interpretation
Donadello, I., Serafini, L., & d'Avila Garcez, A. (2017) · Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2017/221 · Cited by 179 · see citing works
Garcez, A. D., Gori, M., Lamb, L. C., Serafini, L., Spranger, M., & Tran, S. N. (2019) · arXiv · doi:10.48550/arxiv.1905.06088 · Cited by 175 · see citing works
Neural Arithmetic Logic UnitsPreprint
Trask, A., Hill, F., Reed, S., Rae, J. W., Dyer, C., & Blunsom, P. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.00508 · Cited by 153 · see citing works
Abductive learning: towards bridging machine learning and logical reasoning
Zhou, Z. (2019) · Science China Information Sciences · doi:10.1007/s11432-018-9801-4 · Cited by 151 · see citing works
Limits of End-to-End LearningPreprint
Glasmachers, T. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1704.08305 · Cited by 126 · see citing works
(2019) · neural information processing systems · Cited by 104 · see citing works
Self-citationTensorLog: A Probabilistic Database Implemented Using Deep-Learning Infrastructure
Cohen, W. W., Yang, F., & Mazaitis, K. (2020) · Journal of Artificial Intelligence Research · doi:10.1613/jair.1.11944 · Cited by 100 · see citing works
Differentiable Reasoning over a Virtual Knowledge BasePreprint
Dhingra, B., Zaheer, M., Balachandran, V., Neubig, G., Salakhutdinov, R., & Cohen, W. W. (2020) · International Conference on Learning Representations · doi:10.48550/arxiv.2002.10640 · Cited by 91 · see citing works
Stochastic Optimization of Sorting Networks via Continuous RelaxationsPreprint
Grover, A., Wang, E. Z., Zweig, A., & Ermon, S. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1903.08850 · Cited by 79 · see citing works
SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solverPreprint
Wang, P.-W., Donti, P. L., Wilder, B., & Kolter, J. Z. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1905.12149 · Cited by 77 · see citing works
Li, Q., Huang, S., Hong, Y., Chen, Y., Wu, Y. N., & Zhu, S. C. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.06649 · Cited by 46 · see citing works
Differentiable Programs with Neural LibrariesPreprint
Gaunt, A. L., Brockschmidt, M., Kushman, N., & Tarlow, D. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.02109 · Cited by 45 · see citing works
Learning efficient logic programs
Cropper, A., & Muggleton, S. (2018) · Machine Learning · doi:10.1007/s10994-018-5712-6 · Cited by 33 · see citing works
Self-citationFrom Statistical Relational to Neuro-Symbolic Artificial Intelligence
Marra, G. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i20.30294 · Cited by 22 · see citing works
MetaBayes: Bayesian Meta-Interpretative Learning Using Higher-Order Stochastic Refinement
Muggleton, S., Lin, D., Chen, J., & Tamaddoni‐Nezhad, A. (2014) · Lecture notes in computer science · doi:10.1007/978-3-662-44923-3_1 · Cited by 16 · see citing works
Self-citationDai, W.-Z., & Zhou, Z. (2017) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v31i1.11152 · Cited by 13 · see citing works
Self-citationHow Much Can Experimental Cost Be Reduced in Active Learning of Agent Strategies?
Hocquette, C., & Muggleton, S. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-99960-9_3 · Cited by 3 · see citing works
Self-citation
Flagged first, then by citations
1 of the 26 references couldn’t be matched to a known paper.
Research works printed but not linked: 4
- [Bengio, 2017] Yoshua Bengio. The consciousness prior. CoRR , abs/1709.08568, 2017. · arXiv
- [Cropper et al. , 2020] Andrew Cropper, Rolf Morel, and Stephen Muggleton. Learning higher-order logic programs. Maching Learning , 109(7):1289-1322, 2020.
- [Dong et al. , 2019] Honghua Dong, Jiayuan Mao, Tian Lin, Chong Wang, Lihong Li, and Denny Zhou. Neural logic machines. In International Conference on Learning Representations , New Orleans, LA, 2019. OpenReview.
- [Evans and Grefenstette, 2018] Richard Evans and Edward Grefenstette. Learning explanatory rules from noisy data. Journal of Artificial Intelligence Research , 61:1-64, 2018.
Cited by
Sorted by citations
Inductive Logic Programming At 30: A New Introduction
Cropper, A., & Dumančić, S. (2022) · Journal of Artificial Intelligence Research · doi:10.1613/jair.1.13507 · Cited by 83 · see citing works
AI Reasoning in Deep Learning Era: From Symbolic AI to Neural–Symbolic AI
Liang, B., Wang, Y., & Tong, C. (2025) · Mathematics · doi:10.3390/math13111707 · Cited by 79 · see citing works
Inductive logic programming at 30
Cropper, A., Dumančić, S., Evans, R., & Muggleton, S. H. (2022) · Machine Learning · doi:10.1007/s10994-021-06089-1 · Cited by 61 · see citing works
Self-citationRules, L. O.-R., & A, F. A. L. (1999) · Lecture notes in computer science · doi:10.1007/3-540-48751-4 · Cited by 24 · see citing works
FFNSL: Feed-Forward Neural-Symbolic Learner
Cunnington, D., Law, M., Lobo, J., & Russo, A. (2023) · Machine Learning · doi:10.1007/s10994-022-06278-6 · Cited by 20 · see citing works
Explainable Reinforcement and Causal Learning for Improving Trust to 6G Stakeholders
Arana‐Catania, M., Sonee, A., Khan, A.-M., Fatehi, K., Tang, Y., Jin, B., Soligo, A., Boyle, D., Călinescu, R., Yadav, P., et al. (2025) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2025.3563415 · Cited by 18 · see citing works
The Role of Foundation Models in Neuro-Symbolic Learning and Reasoning
Cunnington, D., Law, M., Lobo, J. M., & Russo, A. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-71167-1_5 · Cited by 16 · see citing works
Enabling Knowledge Refinement upon New Concepts in Abductive Learning
Huang, Y., Dai, W.-Z., Jiang, Y. Y., & Zhou, Z. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i7.25959 · Cited by 13 · see citing works
Self-citationNeuro-Symbolic AI + Agent Systems: A First Reflection on Trends, Opportunities and Challenges
Belle, V., Fisher, M., Russo, A., Komendantskaya, E., & Nottle, A. (2023) · AAMAS Workshops · doi:10.1007/978-3-031-56255-6_10 · Cited by 14 · see citing works
Learning differentiable logic programs for abstract visual reasoning
Shindo, H., Pfanschilling, V., Dhami, D. S., & Kersting, K. (2024) · Machine Learning · doi:10.1007/s10994-024-06610-2 · Cited by 16 · see citing works
Enabling Abductive Learning to Exploit Knowledge Graph
Huang, Y., Sun, Z., Li, G., Tian, X., Dai, W.-Z., Hu, W. P., Jiang, Y. Y., & Zhou, Z. (2023) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2023/427 · Cited by 13 · see citing works
Self-citationHypothesizing an algorithm from one example: the role of specificity
FREng, S. H. M. (2023) · Philosophical Transactions of the Royal Society A Mathematical Physical and Engineering Sciences · doi:10.1098/rsta.2022.0046 · Cited by 10 · see citing works
Deep Symbolic Learning: Discovering Symbols and Rules from Perceptions
Daniele, A., Campari, T., Malhotra, S., & Serafini, L. (2023) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2023/400 · Cited by 10 · see citing works
Liu, J., Huang, Z., Zhai, C., & Liu, Q. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i4.25571 · Cited by 9 · see citing works
Nonground Abductive Logic Programming with Probabilistic Integrity Constraints
Bellodi, E., Gavanelli, M., Zese, R., Lamma, E., & Riguzzi, F. (2021) · Theory and Practice of Logic Programming · doi:10.1017/s1471068421000417 · Cited by 8 · see citing works
Out-of-Distribution Generalization by Neural-Symbolic Joint Training
Liu, A., Xu, H., den Broeck, G. V., & Liang, Y. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i10.26444 · Cited by 8 · see citing works
Neuro-Symbolic Learning of Answer Set Programs from Raw Data
Cunnington, D., Law, M., Lobo, J., & Russo, A. (2023) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2023/399 · Cited by 8 · see citing works
Teaching abductive reasoning for use as a problem-solving tool in organic chemistry and beyond
Wackerly, J. W., Wentzel, M. T., & Zingales, S. K. (2024) · Frontiers in Education · doi:10.3389/feduc.2024.1412417 · Cited by 6 · see citing works
Schindler, C., & Rausch, A. (2024) · Electronics · doi:10.3390/electronics13050816 · Cited by 5 · see citing works
Learning MDL Logic Programs from Noisy Data
Hocquette, C., Niskanen, A., Järvisalo, M., & Cropper, A. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i9.28925 · Cited by 6 · see citing works
Deciphering Raw Data in Neuro-Symbolic Learning with Provable Guarantees
Tao, L., Huang, Y., Dai, W.-Z., & Jiang, Y. Y. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i14.29455 · Cited by 4 · see citing works
Self-citationAi, L., Muggleton, S., Liang, S.-S., & Baldwin, G. S. (2025) · Machine Learning · doi:10.1007/s10994-025-06868-0 · Cited by 4 · see citing works
Self-citationLearning Logic Programs by Combining Programs
Cropper, A., & Hocquette, C. (2023) · Frontiers in artificial intelligence and applications · doi:10.3233/faia230309 · Cited by 4 · see citing works
A Primer for Neural Arithmetic Logic ModulesPreprint
Mistry, B., Farrahi, K., & Hare, J. S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2101.09530 · Cited by 3 · see citing works
Neuro-symbolic systems: a perspective of uncertainty management
Li, Z., Yao, Y., Ma, X., & lu, J. (2024) · Scientia Sinica Informationis · doi:10.1360/ssi-2024-0163 · Cited by 2 · see citing works
Embed2Rule Scalable Neuro-Symbolic Learning via Latent Space Weak-Labelling
Aspis, Y., Albinhassan, M., Lobo, J., & Russo, A. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-71167-1_11 · Cited by 2 · see citing works
Generalisation through Negation and Predicate Invention
Cerna, D. M., & Cropper, A. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i9.28915 · Cited by 2 · see citing works
Rule Extraction in Trained Feedforward Deep Neural Networks
Negro, P., & Pons, C. (2024) · International Journal of Artificial Intelligence and Machine Learning · doi:10.4018/ijaiml.347988 · Cited by 2 · see citing works
Symbolic Neural Generation with Applications to Lead Discovery in Drug Design
Srinivasan, A., Dash, T., Baskar, A., Bain, M., Dey, S. K., & Banerjee, M. (2026) · Machine Learning · doi:10.1007/s10994-026-07136-5 · Cited by 1 · see citing works
Rafanelli, A. (2023) · Electronic Proceedings in Theoretical Computer Science · doi:10.4204/eptcs.385.51 · Cited by 1 · see citing works
Abductive learning-guided uncertainty modeling for time series anomaly detection
Zhang, Q., Zhu, M., Li, J., Bao, J., Zhang, D., & Chen, L. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.113878 · Cited by 1 · see citing works
Automated Biodesign Engineering by Abductive Meta-Interpretive LearningPreprint
Dai, W.-Z., Hallett, L., Muggleton, S., & Baldwin, G. S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2105.07758
Self-citationA Comparative Study of Three Neural-Symbolic Approaches to Inductive Logic Programming
Beretta, D., Monica, S., & Bergenti, F. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-15707-3_5
Recent Neural-Symbolic Approaches to ILP Based on Templates
Beretta, D., Monica, S., & Bergenti, F. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-15565-9_5
Understanding SATNet: Constraint Learning and Symbol Grounding
Shakarian, P., Baral, C. R., Simari, G. I., Xi, B., & Pokala, L. (2023) · SpringerBriefs in computer science · doi:10.1007/978-3-031-39179-8_9
Negro, P., & Pons, C. (2024) · Revista Abierta de Informática Aplicada · doi:10.59471/raia2024203
Paraconsistent Abductive Learning for Processing Inconsistent Information
Liu, B., Tanaka, K., & Hossain, M. Z. (2024) · International Conference on Digital Image Computing: Techniques and Applications (DICTA) · doi:10.1109/dicta63115.2024.00100
Data-And Knowledge-Driven Visual Abductive Reasoning
Liang, C., Wang, W., Chen, L., & Yang, Y. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3613712
Pre-Training Meta-Rule Selection Policy for Visual Generative Abductive Learning
Jin, Y., Liu, J., Luo, Z., Peng, Y., Qin, Z., Dai, W.-Z., Ding, Y.-X., & Zhou, K. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-09087-4_11
Self-citationTri-ABL: Enhancing ABL Using Three Learners
Huang, Y. N., He, H.-Y., & Li, M. (2026) · Lecture notes in computer science · doi:10.1007/978-981-92-1462-4_31
Accelerating NeurASP with Vectorization and Caching
Rader, A. P., & Russo, A. (2026) · Theory and Practice of Logic Programming · doi:10.1017/s1471068426100611
Neurosymbolic Learning of Symbolic Automata
Katzouris, N., & Παλιούρας, Γ. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37667-1_31
- Authors
- Dai, W.-Z., & Muggleton, S.
- Affiliations
- Imperial College London (GB)
- Venue
- Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence
- Published
- 2021-08
- Pages
- 1845-1851
- OpenAlex
- W3187230413
- Cited by
- 42
- References
- 26
- Citations by year
- 2026: 4 · 2025: 6 · 2024: 13 · 2023: 11 · 2022: 3 · 2021: 4
- FWCI
- 3.58
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- International Joint Conference on Artificial Intelligence (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Spiral (Imperial College London) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Engineering and Physical Sciences Research Council (EP/R022291/1, EP/R034915/1)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Logic, Reasoning, and Knowledge
- Physical Sciences › Computer Science › Artificial Intelligence › AI-based Problem Solving and Planning
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
- Keywords
- abductive reasoning, knowledge induction, neuro-symbolic learning, predicate invention, prior knowledge, data efficiency
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
