Explainable artificial intelligence: Difference between revisions

Revision as of 06:45, 9 June 2025

Revision as of 23:57, 23 June 2025

Line 80:

[[Marvin Minsky]] et al. raised the issue that AI can function as a form of surveillance, with the biases inherent in surveillance, suggesting HI (Humanistic Intelligence) as a way to create a more fair and balanced "human-in-the-loop" AI.<ref>Minsky, et al., "The Society of Intelligent Veillance" IEEE ISTAS2013, pages 13-17.</ref>

Explainable AI has been recently a new topic researched amongst the context of modern deep learning. Modern complex AI techniques, such as [[deep learning]], are naturally opaque.<ref>{{cite magazine|last1=Mukherjee|first1=Siddhartha|title=A.I. Versus M.D.|url=https://www.newyorker.com/magazine/2017/04/03/ai-versus-md|access-date=30 January 2018|magazine=The New Yorker|date=27 March 2017}}</ref> To address this issue, methods have been developed to make new models more explainable and interpretable.<ref>{{Cite journal|date=2020-07-08|title=Interpretable neural networks based on continuous-valued logic and multicriteria decision operators|journal=Knowledge-Based Systems|language=en|volume=199|pages=105972|doi=10.1016/j.knosys.2020.105972 |arxiv=1910.02486 |issn=0950-7051|doi-access=free|last1=Csiszár|first1=Orsolya|last2=Csiszár|first2=Gábor|last3=Dombi|first3=József}}</ref><ref name="Lipton 31–57"/><ref name="Interpretable machine learning: def"/><ref>{{cite arXiv|last1=Doshi-Velez|first1=Finale|last2=Kim|first2=Been|date=2017-02-27|title=Towards A Rigorous Science of Interpretable Machine Learning|eprint=1702.08608|class=stat.ML}}</ref><ref>{{Cite arXiv |last=Abdollahi, Behnoush, and Olfa Nasraoui.|title=Explainable Restricted Boltzmann Machines for Collaborative Filtering.|eprint=1606.07129|class=stat.ML|year=2016}}</ref><ref>{{Cite book|last1=Dombi|first1=József|last2=Csiszár|first2=Orsolya|series=Studies in Fuzziness and Soft Computing |date=2021|title=Explainable Neural Networks Based on Fuzzy Logic and Multi-criteria Decision Tools|url=https://link.springer.com/book/10.1007/978-3-030-72280-7|volume=408|language=en-gb|doi=10.1007/978-3-030-72280-7|isbn=978-3-030-72279-1|s2cid=233486978|issn=1434-9922}}</ref> This includes layerwise relevance propagation (LRP), a technique for determining which features in a particular input vector contribute most strongly to a neural network's output.<ref name="Bach Binder Montavon Klauschen p=e0130140">{{cite journal|last1=Bach|first1=Sebastian|last2=Binder|first2=Alexander|last3=Montavon|first3=Grégoire|last4=Klauschen|first4=Frederick|last5=Müller|first5=Klaus-Robert|author-link5=Klaus-Robert Müller|last6=Samek|first6=Wojciech|date=2015-07-10|editor-last=Suarez|editor-first=Oscar Deniz|title=On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation|journal=PLOS ONE|volume=10|issue=7|page=e0130140|bibcode=2015PLoSO..1030140B|doi=10.1371/journal.pone.0130140|issn=1932-6203|pmc=4498753|pmid=26161953|doi-access=free}}</ref><ref>{{cite news|url=https://www.theguardian.com/science/2017/nov/05/computer-says-no-why-making-ais-fair-accountable-and-transparent-is-crucial|title=Computer says no: why making AIs fair, accountable and transparent is crucial|last1=Sample|first1=Ian|date=5 November 2017|work=The Guardian|access-date=5 August 2018|language=en}}</ref> Other techniques explain some particular prediction made by a (nonlinear) black-box model, a goal referred to as "local interpretability".<ref>{{Cite journal|last1=Martens|first1=David|last2=Provost|first2=Foster|title=Explaining data-driven document classifications|url=http://pages.stern.nyu.edu/~fprovost/Papers/MartensProvost_Explaining.pdf|journal=MIS Quarterly|year=2014|volume=38|pages=73–99|doi=10.25300/MISQ/2014/38.1.04|s2cid=14238842}}</ref><ref>{{Cite journal|title="Why Should I Trust You?" {{!}} Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining|language=EN|doi=10.1145/2939672.2939778|s2cid=13029170}}</ref><ref>{{Citation|last1=Lundberg|first1=Scott M|title=A Unified Approach to Interpreting Model Predictions|date=2017|url=http://papers.nips.cc/paper/7062-a-unified-approach-to-interpreting-model-predictions.pdf|work=Advances in Neural Information Processing Systems 30|pages=4765–4774|editor-last=Guyon|editor-first=I.|publisher=Curran Associates, Inc.|access-date=2020-03-13|last2=Lee|first2=Su-In|author2-link=Su-In Lee|editor2-last=Luxburg|editor2-first=U. V.|editor3-last=Bengio|editor3-first=S.|editor4-last=Wallach|editor4-first=H.|arxiv=1705.07874}}</ref><ref>{{Cite journal|last1=Carter|first1=Brandon|last2=Mueller|first2=Jonas|last3=Jain|first3=Siddhartha|last4=Gifford|first4=David|date=2019-04-11|title=What made you do this? Understanding black-box decisions with sufficient input subsets|url=http://proceedings.mlr.press/v89/carter19a.html|journal=The 22nd International Conference on Artificial Intelligence and Statistics|language=en|pages=567–576}}</ref><ref>{{Cite journal|last1=Shrikumar|first1=Avanti|last2=Greenside|first2=Peyton|last3=Kundaje|first3=Anshul|date=2017-07-17|title=Learning Important Features Through Propagating Activation Differences|url=http://proceedings.mlr.press/v70/shrikumar17a.html|journal=International Conference on Machine Learning|language=en|pages=3145–3153}}</ref><ref>{{Cite journal|url=https://dl.acm.org/doi/abs/10.5555/3305890.3306024|title=Axiomatic attribution for deep networks {{!}} Proceedings of the 34th International Conference on Machine Learning - Volume 70|website=dl.acm.org|series=Icml'17|date=6 August 2017|pages=3319–3328|language=EN|access-date=2020-03-13}}</ref> We still today cannot explain the output of today's DNNs without the new explanatory mechanisms, we also can't by the neural network, or external explanatory components <ref>{{Citation |last1=Xu |first1=Feiyu |title=Explainable AI: A Brief Survey on History, Research Areas, Approaches and Challenges |date=2019 |work=Natural Language Processing and Chinese Computing |volume=11839 |pages=563–574 |editor-last=Tang |editor-first=Jie |url=http://link.springer.com/10.1007/978-3-030-32236-6_51 |access-date=2024-12-03 |place=Cham |publisher=Springer International Publishing |language=en |doi=10.1007/978-3-030-32236-6_51 |isbn=978-3-030-32235-9 |last2=Uszkoreit |first2=Hans |last3=Du |first3=Yangzhou |last4=Fan |first4=Wei |last5=Zhao |first5=Dongyan |last6=Zhu |first6=Jun |editor2-last=Kan |editor2-first=Min-Yen |editor3-last=Zhao |editor3-first=Dongyan |editor4-last=Li |editor4-first=Sujian|url-access=subscription }}</ref> There is also research on whether the concepts of local interpretability can be applied to a remote context, where a model is operated by a third-party.<ref>{{Cite journal|last1=Le Merrer|first1=Erwan|last2=Trédan|first2=Gilles|date=September 2020|title=Remote explainability faces the bouncer problem|url=https://www.nature.com/articles/s42256-020-0216-z|journal=Nature Machine Intelligence|language=en|volume=2|issue=9|pages=529–539|doi=10.1038/s42256-020-0216-z|issn=2522-5839|arxiv=1910.01432|s2cid=225207140}}</ref><ref>{{Cite journal |last1=Aivodji |first1=Ulrich |last2=Arai |first2=Hiromi |last3=Fortineau |first3=Olivier |last4=Gambs |first4=Sébastien |last5=Hara |first5=Satoshi |last6=Tapp |first6=Alain |date=2019-05-24 |title=Fairwashing: the risk of rationalization |url=http://proceedings.mlr.press/v97/aivodji19a.html |journal=International Conference on Machine Learning |language=en |publisher=PMLR |pages=161–170 |arxiv=1901.09749}}</ref>

There has been work on making glass-box models which are more transparent to inspection.<ref name=":6"/><ref>{{cite journal |last1=Singh |first1=Chandan |last2=Nasseri |first2=Keyan |last3=Tan |first3=Yan Shuo |last4=Tang |first4=Tiffany |last5=Yu |first5=Bin |title=imodels: a python package for fitting interpretable models |journal=Journal of Open Source Software |date=4 May 2021 |volume=6 |issue=61 |pages=3192 |doi=10.21105/joss.03192 |bibcode=2021JOSS....6.3192S |s2cid=235529515 |url=https://joss.theoj.org/papers/10.21105/joss.03192 |language=en |issn=2475-9066|doi-access=free }}</ref> This includes [[decision tree]]s,<ref>{{Cite journal|last1=Vidal|first1=Thibaut|last2=Schiffer|first2=Maximilian|date=2020|title=Born-Again Tree Ensembles|url=http://proceedings.mlr.press/v119/vidal20a.html|journal=International Conference on Machine Learning|language=en|publisher=PMLR|volume=119|pages=9743–9753|arxiv=2003.11132}}</ref> [[Bayesian network]]s, sparse [[linear model]]s,<ref>{{cite journal |last1=Ustun |first1=Berk |last2=Rudin |first2=Cynthia |title=Supersparse linear integer models for optimized medical scoring systems |journal=Machine Learning |date=1 March 2016 |volume=102 |issue=3 |pages=349–391 |doi=10.1007/s10994-015-5528-6 |s2cid=207211836 |url=https://link.springer.com/article/10.1007/s10994-015-5528-6 |language=en |issn=1573-0565|arxiv=1502.04269 }}</ref> and more.<ref>Bostrom, N., & Yudkowsky, E. (2014). [https://intelligence.org/files/EthicsofAI.pdf The ethics of artificial intelligence]. ''The Cambridge Handbook of Artificial Intelligence'', 316-334.</ref> The [[ACM Conference on Fairness, Accountability, and Transparency|Association for Computing Machinery Conference on Fairness, Accountability, and Transparency (ACM FAccT)]] was established in 2018 to study transparency and explainability in the context of socio-technical systems, many of which include artificial intelligence.<ref name="FAT* conference">{{cite web | url=https://fatconference.org/ | title=FAT* Conference }}</ref><ref>{{cite news |title=Computer programs recognise white men better than black women |url=https://www.economist.com/science-and-technology/2018/02/15/computer-programs-recognise-white-men-better-than-black-women |access-date=5 August 2018 |newspaper=The Economist |date=2018 |language=en}}</ref>

Line 117:

This outcome was especially true for decisions that impacted the end user in a significant way, such as graduate school admissions. Participants judged algorithms to be too inflexible and unforgiving in comparison to human decision-makers; instead of rigidly adhering to a set of rules, humans are able to consider exceptional cases as well as appeals to their initial decision.<ref name="Explaining Decision-Making Algorithms through UI" /> For such decisions, explainability will not necessarily cause end users to accept the use of decision-making algorithms. We will need to either turn to another method to increase trust and acceptance of decision-making algorithms, or question the need to rely solely on AI for such impactful decisions in the first place.

However, some emphasize that the purpose of explainability of artificial intelligence is not to merely increase users' trust in the system's decisions, but to calibrate the users' level of trust to the correct level.<ref>{{Cite journal |last1=Liel |first1=Yotam |last2=Zalmanson |first2=Lior |date=August 2023 |title=Turning Off Your Better Judgment – Conformity to Algorithmic Recommendations |url=http://journals.aom.org/doi/full/10.5465/AMPROC.2023.277bp |journal=Academy of Management Proceedings |language=en |volume=2023 |issue=1 |doi=10.5465/AMPROC.2023.277bp |s2cid=260212999 |issn=0065-0668|url-access=subscription }}</ref> According to this principle, too much or too little user trust in the AI system will harm the overall performance of the human-system unit. When the trust is excessive, the users are not critical of possible mistakes of the system and when the users do not have enough trust in the system, they will not exhaust the benefits inherent in it.

== Criticism ==

Line 127:

== Explainability in social choice ==

Explainability was studied also in [[social choice theory]]. Social choice theory aims at finding solutions to social decision problems, that are based on well-established axioms. [[Ariel D. Procaccia]]<ref>{{Citation |last=Procaccia |first=Ariel D. |title=Axioms Should Explain Solutions |date=2019 |work=The Future of Economic Design: The Continuing Development of a Field as Envisioned by Its Researchers |series=Studies in Economic Design |pages=195–199 |editor-last=Laslier |editor-first=Jean-François |url=https://doi.org/10.1007/978-3-030-18050-8_27 |access-date=2024-03-21 |place=Cham |publisher=Springer International Publishing |language=en |doi=10.1007/978-3-030-18050-8_27 |isbn=978-3-030-18050-8 |editor2-last=Moulin |editor2-first=Hervé |editor3-last=Sanver |editor3-first=M. Remzi |editor4-last=Zwicker |editor4-first=William S.|url-access=subscription }}</ref> explains that these axioms can be used to construct convincing explanations to the solutions. This principle has been used to construct explanations in various subfields of social choice.

=== Voting ===