Beneficial and harmful explanatory machine learning
Open accessThe DOI matches its Crossref record; 14 of the 14 references it cites were found in a registry or an index.
Abstract
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.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Mastering the game of Go with deep neural networks and tree search
Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., van den Driessche, G., Schrittwieser, J., Antonoglou, I., Panneershelvam, V., Lanctot, M., et al. (2016) · Nature · doi:10.1038/nature16961 · Cited by 15,612
Bengio, Y., Louradour, J., Collobert, R., & Weston, J. (2009) · Proceedings of the 26th Annual International Conference on Machine Learning · doi:10.1145/1553374.1553380 · Cited by 5,366 · cited 1×
“Therefore, learning might be organised incrementally, guided by a curriculum [6,52].”
BackgroundExplanation in artificial intelligence: Insights from the social sciences
Miller, T. (2019) · Artificial Intelligence · doi:10.1016/j.artint.2018.07.007 · Cited by 5,302
Quinlan, J. R. (1987) · International Journal of Man-Machine Studies · doi:10.1016/s0020-7373(87)80053-6 · Cited by 2,553
Carpenter, P. A., Just, M. A., & Shell, P. (1990) · Psychological Review · doi:10.1037/0033-295x.97.3.404 · Cited by 1,648
Muggleton, S. (1991) · New Generation Computing · doi:10.1007/bf03037089 · Cited by 1,044
Self-citationAleven, V. A. W. M. M., & Koedinger, K. R. (2002) · Cognitive Science · doi:10.1207/s15516709cog2602_1 · Cited by 895
Memory for goals: an activation‐based model
Altmann, E. M., & Trafton, J. G. (2002) · Cognitive Science · doi:10.1207/s15516709cog2601_2 · Cited by 795 · cited 2×
“If the strategy involves to keep in mind a stack of open sub-goals – as for example the Tower of Hanoi [2,46] – explanations might become more complex than figuring out the action sequence.”
BackgroundRelational Reinforcement Learning
Džeroski, S., De Raedt, L., & Driessens, K. (2001) · Machine Learning · doi:10.1023/a:1007694015589 · Cited by 371
Miller, T., Howe, P., & Sonenberg, L. (2017) · arXiv · doi:10.48550/arxiv.1712.00547 · Cited by 218
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
Self-citationMeta-interpretive learning: application to grammatical inference
Muggleton, S. H., Lin, D., Pahlavi, N., & Tamaddoni-Nezhad, A. (2014) · Machine Learning · doi:10.1007/s10994-013-5358-3 · Cited by 144
Self-citationBias reformulation for one-shot function induction
Dianhuan, L., Eyal, D., Kevin, E., Joshua, T., & Stephen, M. (2014) · Frontiers in Artificial Intelligence and Applications · doi:10.3233/978-1-61499-419-0-525 · Cited by 53
Behavioural Cloning: Phenomena, Results and Problems
Bratko, I., Urbančič, T., & Sammut, C. (1995) · IFAC Proceedings Volumes · doi:10.1016/s1474-6670(17)46716-4 · Cited by 46
Cited by
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
“Recent work (Muggleton et al., 2018; Ai et al., 2021) evaluates the comprehensibility of ILP hypotheses using Michie’s (1988) framework of ultra-strong ML, where a learned hypothesis is expected to not only be accurate but to also demonstrably improve the performance of a human being provided with the learned hypothes…”
MethodInductive 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
Self-citationA Critical Review of Inductive Logic Programming Techniques for Explainable AI
Zhang, Z., Yilmaz, L., & Liu, B. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3246980 · Cited by 25
Toward Accountable and Explainable Artificial Intelligence Part Two: The Framework Implementation
Vice, J., & Khan, M. M. (2022) · IEEE Access · doi:10.1109/access.2022.3163523 · Cited by 11
Metadata
- Authors
- Ai, L., Muggleton, S. H., Hocquette, C., Gromowski, M., & Schmid, U.
- Venue
- Machine Learning
- Published
- 2021-03-11
- Volume
- 110
- Issue
- 4
- Pages
- 695-721
- ISSN
- 0885-6125 (ISSN-L)
- 1573-0565
- OpenAlex
- W3139019272
- Cited by
- 34
- References
- 14
- FWCI
- 2.47
- Type
- journal article
- Language
- en
- Open access
- hybrid
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.