pith. sign in

arxiv: 2405.10729 · v2 · submitted 2024-05-17 · 💻 cs.AI

Contestable AI needs Computational Argumentation

Francesco Leofante , Hamed Ayoobi , Adam Dejl , Gabriel Freedman , Deniz Gorur , Junqi Jiang , Guilherme Paulino-Passos , Antonio Rago
show 5 more authors
Anna Rapberger Fabrizio Russo Xiang Yin Dekai Zhang Francesca Toni
This is my paper

Pith reviewed 2026-05-24 01:17 UTC · model grok-4.3

classification 💻 cs.AI
keywords contestabilitycomputational argumentationexplainable AIdynamic decision makingAI regulationGDPRautomated decisions
0
0 comments X

The pith

Contestable AI requires dynamic explainability and decision revision that computational argumentation can supply.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

AI guidelines and regulations such as the GDPR call for decisions that people can challenge and have revised. Most current AI systems produce fixed outputs without ongoing interaction. The paper argues that contestability therefore demands dynamic processes in which machines explain their reasoning, evaluate contestation grounds from humans or other machines, and revise decisions when issues are raised. It positions computational argumentation as the mechanism suited to structure these interactions and drive the needed shift away from static designs.

Core claim

The paper claims that contestable AI requires dynamic human-machine and machine-machine processes for progressive explanation, assessment of contestation grounds, and revision of decisions, and that computational argumentation is ideally suited to support the radical rethinking of AI systems away from static approaches toward these interactive capabilities.

What carries the argument

Computational argumentation as a framework for structured reasoning that enables interactive explanation, contest assessment, and decision revision.

If this is right

  • AI systems become able to interact with users or other systems to explain outputs step by step.
  • Contestation grounds can be assessed and used to trigger actual revisions in the decision process.
  • Machine-to-machine contestation becomes feasible in addition to human challenges.
  • AI design must move from one-shot static models to revisable, interactive architectures.
  • Regulatory compliance for automated decisions shifts to rely on argumentation-based dynamics.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • This view could lead to new evaluation metrics that test whether an AI actually revises outputs after contestation rather than only explaining them.
  • It connects to multi-agent AI settings where agents use arguments to negotiate or override decisions.
  • Practical implementations might combine argumentation layers with existing machine learning models to add contestability without full redesign.
  • The approach raises questions about how to scale argument-based interactions to high-volume decision systems.

Load-bearing premise

Static AI outputs cannot meet contestability requirements and dynamic interaction is both necessary and sufficient to satisfy them.

What would settle it

A static AI system that fully satisfies GDPR-style contestability rules without any capacity for ongoing explanation or decision revision.

Figures

Figures reproduced from arXiv: 2405.10729 by Adam Dejl, Anna Rapberger, Antonio Rago, Dekai Zhang, Deniz Gorur, Fabrizio Russo, Francesca Toni, Francesco Leofante, Gabriel Freedman, Guilherme Paulino-Passos, Hamed Ayoobi, Junqi Jiang, Xiang Yin.

Figure 1
Figure 1. Figure 1: An abstract view of AI contestability: the contested ADS [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
read the original abstract

AI has become pervasive in recent years, but state-of-the-art approaches predominantly neglect the need for AI systems to be contestable. Instead, contestability is advocated by AI guidelines (e.g. by the OECD) and regulation of automated decision-making (e.g. GDPR). In this position paper we explore how contestability can be achieved computationally in and for AI. We argue that contestable AI requires dynamic (human-machine and/or machine-machine) explainability and decision-making processes, whereby machines can (i) interact with humans and/or other machines to progressively explain their outputs and/or their reasoning as well as assess grounds for contestation provided by these humans and/or other machines, and (ii) revise their decision-making processes to redress any issues successfully raised during contestation. Given that much of the current AI landscape is tailored to static AIs, the need to accommodate contestability will require a radical rethinking, that, we argue, computational argumentation is ideally suited to support.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. This position paper argues that contestable AI requires dynamic (human-machine and/or machine-machine) explainability and decision-making processes in which systems can interact to explain outputs, assess contestation grounds, and revise decisions accordingly. It claims that the current landscape of predominantly static AI systems cannot accommodate these requirements without radical rethinking, and that computational argumentation is ideally suited to provide the necessary support, aligning with guidelines such as those from the OECD and regulations like the GDPR.

Significance. If the position holds, it would usefully connect regulatory demands for contestability with computational argumentation theory, potentially informing the design of more adaptable AI systems. The paper gives credit to the alignment with existing regulatory documents and prior literature on argumentation without introducing new formal machinery or empirical tests.

major comments (2)
  1. [Abstract] Abstract: The central claim that static AI approaches 'cannot' accommodate contestability (and thus require 'radical rethinking') is load-bearing for the advocacy but is asserted without a concrete analysis of why existing static explainability techniques (e.g., feature attribution or counterfactual methods) fail to satisfy the cited regulatory requirements for contestation.
  2. [Abstract] Abstract: The sufficiency claim that dynamic processes are both necessary and sufficient for contestability rests on an untested premise about regulatory interpretation; the paper does not demonstrate that argumentation frameworks can deliver the required revision properties in a way that static methods cannot.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments on our position paper. We address each major comment below, focusing on the conceptual nature of our arguments and indicating where revisions to the abstract will improve clarity.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central claim that static AI approaches 'cannot' accommodate contestability (and thus require 'radical rethinking') is load-bearing for the advocacy but is asserted without a concrete analysis of why existing static explainability techniques (e.g., feature attribution or counterfactual methods) fail to satisfy the cited regulatory requirements for contestation.

    Authors: The abstract summarizes our position that predominantly static AI systems require radical rethinking to meet contestability demands under regulations such as the GDPR. The full manuscript provides a conceptual discussion of how static techniques like feature attribution deliver one-directional outputs that do not support the interactive assessment of contestation grounds or subsequent decision revision. We agree the abstract would benefit from a brief signal of this distinction and will revise it accordingly to reference the regulatory emphasis on meaningful contestation. revision: yes

  2. Referee: [Abstract] Abstract: The sufficiency claim that dynamic processes are both necessary and sufficient for contestability rests on an untested premise about regulatory interpretation; the paper does not demonstrate that argumentation frameworks can deliver the required revision properties in a way that static methods cannot.

    Authors: As a position paper we advance a conceptual argument that contestability, as framed by the cited guidelines and regulations, requires dynamic interaction and revision capabilities; we do not claim empirical demonstration or formal proof of necessity and sufficiency. Computational argumentation is presented as well-suited for these properties due to its support for dialectical exchange and argument-based updates. We will revise the abstract to clarify the advocacy nature of the claim and that implementation and validation remain for future work. revision: partial

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

This is a position paper whose central claim is an advocacy argument rather than a technical derivation: contestability requires dynamic human/machine processes, static AI cannot accommodate this without radical change, and computational argumentation is suited to support it. The argument draws on external sources (OECD guidelines, GDPR) and prior literature on argumentation without any equations, fitted parameters, self-definitional reductions, or load-bearing self-citation chains that reduce the recommendation to its own inputs by construction. No patterns from the enumerated list apply, and the paper is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The paper relies on domain assumptions about regulatory requirements and the limitations of static AI without introducing new free parameters, invented entities, or formal axioms beyond standard background in AI ethics.

axioms (2)
  • domain assumption AI systems must support contestation to comply with guidelines such as OECD and regulations such as GDPR
    Invoked in the opening paragraph as the motivation for the position.
  • domain assumption Static AI cannot accommodate the required dynamic explainability and revision processes
    Stated as the reason a radical rethinking is needed.

pith-pipeline@v0.9.0 · 5733 in / 1271 out tokens · 29219 ms · 2026-05-24T01:17:05.269790+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

124 extracted references · 124 canonical work pages · 1 internal anchor

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION fin.entry add.period write newline FUNCTION new.block output.state before.all = 'skip after.block 'output.state := if FUNCTION new.sentence output.state after.block = 'skip output.state before.all = 'skip after.sentence 'output.state := if if FUNCTION not #0 #1 if FUNCTION and 'skip pop #0 if FUNCTIO...

    work page

  2. [2]

    Albini, E.; Lertvittayakumjorn, P.; Rago, A.; and Toni, F. 2020. DAX: deep argumentative explanation for neural networks. CoRR abs/2012.05766

    work page arXiv 2020

  3. [3]

    Alfano, G.; Calautti, M.; Greco, S.; Parisi, F.; and Trubitsyna, I. 2023. Explainable acceptance in probabilistic and incomplete abstract argumentation frameworks. Artif. Intell. 323:103967

    work page 2023

  4. [4]

    Alfrink, K.; Keller, I.; Kortuem, G.; and Doorn, N. 2022. Contestable ai by design: Towards a framework. Minds and Machines 1--27

    work page 2022

  5. [5]

    Alfrink, K.; Keller, I.; Doorn, N.; and Kortuem, G. 2023. Contestable camera cars: A speculative design exploration of public AI that is open and responsive to dispute. In Proc. of CHI , 8:1--8:16

    work page 2023

  6. [6]

    Almada, M. 2019. Human intervention in automated decision-making: Toward the construction of contestable systems. In Proc. of ICAIL , 2--11

    work page 2019

  7. [7]

    Almog, S., and Kalech, M. 2023. Diagnosis for post concept drift decision trees repair. In Proc. of KR , 23--33

    work page 2023

  8. [8]

    Amgoud, L.; Ben-Naim, J.; and Vesic, S. 2017. Measuring the intensity of attacks in argumentation graphs with shapley value. In Proc. of IJCAI , 63--69

    work page 2017

  9. [9]

    Arioua, A.; Tamani, N.; and Croitoru, M. 2015. Query answering explanation in inconsistent datalog +/- knowledge bases. In Proc. of DEXA , 203--219

    work page 2015

  10. [10]

    R.; Thimm, M.; and Villata, S

    Atkinson, K.; Baroni, P.; Giacomin, M.; Hunter, A.; Prakken, H.; Reed, C.; Simari, G. R.; Thimm, M.; and Villata, S. 2017. Towards artificial argumentation. AI Magazine 38(3):25--36

    work page 2017

  11. [11]

    Ayoobi, H.; Potyka, N.; and Toni, F. 2023. SpArX : Sparse argumentative explanations for neural networks. In Proc. of ECAI , volume 372, 149--156

    work page 2023

  12. [12]

    Baroni, P.; Gabbay, D.; Giacomin, M.; and van der Torre, L., eds. 2018. Handbook of Formal Argumentation . College Publications

    work page 2018

  13. [13]

    Baumann, R., and Brewka, G. 2015. AGM meets abstract argumentation: Expansion and revision for dung frameworks. In Proc. of IJCAI , 2734--2740

    work page 2015

  14. [14]

    Baumann, R.; Doutre, S.; Mailly, J.; and Wallner, J. P. 2021. Enforcement in formal argumentation. FLAP 8(6):1623--1678

    work page 2021

  15. [15]

    M.; and Rodrigues, O

    Baumann, R.; Gabbay, D. M.; and Rodrigues, O. 2020. Forgetting an argument. In Proc. of AAAI , 2750--2757

    work page 2020

  16. [16]

    Baumeister, D.; J \" a rvisalo, M.; Neugebauer, D.; Niskanen, A.; and Rothe, J. 2021. Acceptance in incomplete argumentation frameworks. Artif. Intell. 295:103470

    work page 2021

  17. [17]

    Berthold, M.; Rapberger, A.; and Ulbricht, M. 2023. Forgetting aspects in assumption-based argumentation. In Proc. of KR , 86--96

    work page 2023

  18. [18]

    Bills, S.; Cammarata, N.; Mossing, D.; Tillman, H.; Gao, L.; Goh, G.; Sutskever, I.; Leike, J.; Wu, J.; and Saunders, W. 2023. Language models can explain neurons in language models. https://openaipublic. blob. core. windows. net/neuron-explainer/paper/index. [Accessed: 04.08.23]

    work page 2023

  19. [19]

    Black, E., and Hunter, A. 2007. A generative inquiry dialogue system. In Proc. of (AAMAS , 241

    work page 2007

  20. [20]

    Booth, R.; Kaci, S.; Rienstra, T.; and van der Torre, L. W. N. 2013. A logical theory about dynamics in abstract argumentation. In Proc. of SUM , volume 8078, 148--161

    work page 2013

  21. [21]

    Borg, A., and Bex, F. 2020. Explaining arguments at the dutch national police. In Proc. of AICOL , 183--197

    work page 2020

  22. [22]

    E.; Bud \' a n, M

    Briguez, C. E.; Bud \' a n, M. C.; Deagustini, C. A. D.; Maguitman, A. G.; Capobianco, M.; and Simari, G. R. 2014. Argument-based mixed recommenders and their application to movie suggestion. Expert Syst. Appl. 41(14):6467--6482

    work page 2014

  23. [23]

    Calegari, R.; Omicini, A.; Pisano, G.; and Sartor, G. 2022. Arg2P : an argumentation framework for explainable intelligent systems. J. Log. Comput. 32(2):369--401

    work page 2022

  24. [24]

    Calegari, R.; Riveret, R.; and Sartor, G. 2021. The burden of persuasion in structured argumentation. In Proc. of ICAIL , 180--184

    work page 2021

  25. [25]

    Cao, L. 2022. Ai in finance: challenges, techniques, and opportunities. ACM Computing Surveys (CSUR) 55(3):1--38

    work page 2022

  26. [26]

    Carey, R., and Everitt, T. 2023. Human control: Definitions and algorithms. In Proc. of UAI , 271--281

    work page 2023

  27. [27]

    Carstens, L., and Toni, F. 2017. Using argumentation to improve classification in natural language problems. ACM Trans. Internet Techn. 17(3):30:1--30:23

    work page 2017

  28. [28]

    D.; and Lagasquie - Schiex, M

    Cayrol, C.; de Saint - Cyr, F. D.; and Lagasquie - Schiex, M. 2010. Change in abstract argumentation frameworks: Adding an argument. J. Artif. Intell. Res. 38:49--84

    work page 2010

  29. [29]

    Cocarascu, O.; Stylianou, A.; Cyras, K.; and Toni, F. 2020. Data-empowered argumentation for dialectically explainable predictions. In Proc. of ECAI , 2449--2456

    work page 2020

  30. [30]

    C yras, K.; Birch, D.; Guo, Y.; Toni, F.; Dulay, R.; Turvey, S.; Greenberg, D.; and Hapuarachchi, T. 2019a. Explanations by arbitrated argumentative dispute. Expert Systems with Applications 127:141--156

    work page

  31. [31]

    Cyras, K.; Letsios, D.; Misener, R.; and Toni, F. 2019b. Argumentation for explainable scheduling. In Proc. of AAAI , 2752--2759

    work page

  32. [32]

    Cyras, K.; Rago, A.; Albini, E.; Baroni, P.; and Toni, F. 2021. Argumentative XAI: A survey. In Proc. of IJCAI , 4392--4399

    work page 2021

  33. [33]

    Darwiche, A. 2020. Three modern roles for logic in AI . In Proc. of PODS 2020 , 229--243. ACM

    work page 2020

  34. [34]

    D.; Bonzon, E.; and Maudet, N

    de Tarl \' e , L. D.; Bonzon, E.; and Maudet, N. 2022. Multiagent dynamics of gradual argumentation semantics. In Proc. of AAMAS , 363--371

    work page 2022

  35. [35]

    Donadello, I.; Hunter, A.; Teso, S.; and Dragoni, M. 2022. Machine learning for utility prediction in argument-based computational persuasion. In Proc. of AAAI , 5592--5599

    work page 2022

  36. [36]

    Doutre, S., and Mailly, J. 2018. Constraints and changes: A survey of abstract argumentation dynamics. Argument Comput. 9(3):223--248

    work page 2018

  37. [37]

    J.; Samek, W.; and Lapuschkin, S

    Dreyer, M.; Pahde, F.; Anders, C. J.; Samek, W.; and Lapuschkin, S. 2024. From Hope to Safety : Unlearning Biases of Deep Models via Gradient Penalization in Latent Space . Proc. of AAAI 21046--21054

    work page 2024

  38. [38]

    A.; Kern - Isberner, G.; and Simari, G

    Falappa, M. A.; Kern - Isberner, G.; and Simari, G. R. 2009. Belief revision and argumentation theory. In Argumentation in Artificial Intelligence . Springer. 341--360

    work page 2009

  39. [39]

    Fan, X., and Toni, F. 2012. Mechanism design for argumentation-based persuasion. In Proc. of COMMA , 322--333

    work page 2012

  40. [40]

    Fan, X., and Toni, F. 2015a. Mechanism design for argumentation-based information-seeking and inquiry. In Proc. of PRIMA , 519--527

    work page

  41. [41]

    Fan, X., and Toni, F. 2015b. On computing explanations in argumentation. In Proc. of AAAI , 1496--1502

    work page

  42. [42]

    Fan, X. 2018. On generating explainable plans with assumption-based argumentation. In Proc. of PRIMA , 344--361

    work page 2018

  43. [43]

    Ferreira, J.; de Sousa Ribeiro, M.; Gon c alves, R.; and Leite, J. 2022. Looking inside the black-box: Logic-based explanations for neural networks. In Proc. of KR

    work page 2022

  44. [44]

    Freedman, G.; Dejl, A.; Gorur, D.; Yin, X.; Rago, A.; and Toni, F. 2024. Argumentative large language models for explainable and contestable decision-making. CoRR abs/2405.02079

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  45. [45]

    Gabriel, I. 2020. Artificial intelligence, values, and alignment. Minds Mach. 30(3):411--437

    work page 2020

  46. [46]

    V.; Zhang, Y.; Bellamy, R.; and Mueller, K

    Ghai, B.; Liao, Q. V.; Zhang, Y.; Bellamy, R.; and Mueller, K. 2021. Explainable active learning (xal) toward ai explanations as interfaces for machine teachers. Proc. of HCI 1--28

    work page 2021

  47. [47]

    Guidotti, R.; Monreale, A.; Giannotti, F.; Pedreschi, D.; Ruggieri, S.; and Turini, F. 2019. Factual and counterfactual explanations for black box decision making. IEEE Intell. Syst. 34(6):14--23

    work page 2019

  48. [48]

    Hadfield - Menell, D.; Russell, S.; Abbeel, P.; and Dragan, A. D. 2016. Cooperative inverse reinforcement learning. In Proc. of NeurIPS , 3909--3917

    work page 2016

  49. [49]

    Henin, C., and M \' e tayer, D. L. 2022. Beyond explainability: justifiability and contestability of algorithmic decision systems. AI Soc. 37(4):1397--1410

    work page 2022

  50. [50]

    Henriksen, P.; Leofante, F.; and Lomuscio, A. 2022. Repairing misclassifications in neural networks using limited data. In Proc. of SAC , 1031--1038

    work page 2022

  51. [51]

    Hicks, A. 2022. Transparency, compliance, and contestability when code is(n't) law. In Proc. of NSPW , 130--142

    work page 2022

  52. [52]

    S.; Imel, Z

    Hirsch, T.; Merced, K.; Narayanan, S. S.; Imel, Z. E.; and Atkins, D. C. 2017. Designing contestability: Interaction design, machine learning, and mental health. In Proc. of DIS , 95--99

    work page 2017

  53. [53]

    Hunter, A. 2018. Towards a framework for computational persuasion with applications in behaviour change. Argument Comput. 9(1):15--40

    work page 2018

  54. [54]

    Ignatiev, A.; Narodytska, N.; and Marques - Silva, J. 2019. Abduction-based explanations for machine learning models. In Proc. of AAAI , 1511--1519. AAAI Press

    work page 2019

  55. [55]

    Jiang, J.; Leofante, F.; Rago, A.; and Toni, F. 2023. Formalising the robustness of counterfactual explanations for neural networks. In Proc. of AAAI , 14901--14909

    work page 2023

  56. [56]

    Jiang, J.; Leofante, F.; Rago, A.; and Toni, F. 2024a. Recourse under model multiplicity via argumentative ensembling. In Proc. of AAMAS , 954--963

    work page

  57. [57]

    Jiang, J.; Leofante, F.; Rago, A.; and Toni, F. 2024b. Robust counterfactual explanations in machine learning: A survey. In Proc. of IJCAI , 8086--8094

    work page

  58. [58]

    M., and Keane, M

    Kenny, E. M., and Keane, M. T. 2019. Twin-systems to explain artificial neural networks using case-based reasoning: Comparative tests of feature-weighting methods in ANN-CBR twins for XAI . In Proc. of IJCAI , 2708--2715

    work page 2019

  59. [59]

    J.; Wexler, J.; Vi \' e gas, F

    Kim, B.; Wattenberg, M.; Gilmer, J.; Cai, C. J.; Wexler, J.; Vi \' e gas, F. B.; and Sayres, R. 2018. Interpretability beyond feature attribution: Quantitative testing with concept activation vectors (TCAV) . In Proc. of ICML , 2673--2682

    work page 2018

  60. [60]

    N.; Kohli, N.; and Mulligan, D

    Kluttz, D. N.; Kohli, N.; and Mulligan, D. K. 2022. Shaping our tools: Contestability as a means to promote responsible algorithmic decision making in the professions. In Ethics of Data and Analytics . 420--428

    work page 2022

  61. [61]

    Kontarinis, D., and Toni, F. 2015. Identifying malicious behavior in multi-party bipolar argumentation debates. In Proc. of EUMAS , 267--278

    work page 2015

  62. [62]

    D.; and Lomuscio, A

    Kouvaros, P.; Leofante, F.; Edwards, B.; Chung, C.; Margineantu, D. D.; and Lomuscio, A. 2023. Verification of semantic key point detection for aircraft pose estimation. In Proc. of KR , 757--762

    work page 2023

  63. [63]

    Lakkaraju, H.; Kamar, E.; Caruana, R.; and Leskovec, J. 2019. Faithful and customizable explanations of black box models. In Proc. of AIES , 131--138

    work page 2019

  64. [64]

    Lawrence, J.; Visser, J.; and Reed, C. 2023. Translational argument technology: Engineering a step change in the argument web. J. Web Semant. 77:100786

    work page 2023

  65. [65]

    Leofante, F., and Lomuscio, A. 2023. Robust explanations for human-neural multi-agent systems with formal verification. In Proc. of EUMAS , 244--262

    work page 2023

  66. [66]

    Leofante, F., and Potyka, N. 2024. Promoting counterfactual robustness through diversity. In Proc. of AAAI , 21322--21330

    work page 2024

  67. [67]

    Leofante, F.; Botoeva, E.; and Rajani, V. 2023. Counterfactual explanations and model multiplicity: a relational verification view. In Proc. of KR , 763--768

    work page 2023

  68. [68]

    Lertvittayakumjorn, P., and Toni, F. 2021. Explanation-based human debugging of NLP models: A survey. Trans. Assoc. Comput. Linguistics 9:1508--1528

    work page 2021

  69. [69]

    Lertvittayakumjorn, P.; Specia, L.; and Toni, F. 2020. FIND: human-in-the-loop debugging deep text classifiers. In Proc. of EMNLP , 332--348

    work page 2020

  70. [70]

    M., and Lee, S.-I

    Lundberg, S. M., and Lee, S.-I. 2017. A unified approach to interpreting model predictions. In Proc. of NeurIPS , 4765--4774

    work page 2017

  71. [71]

    Lyons, H.; Velloso, E.; and Miller, T. 2021. Conceptualising contestability: Perspectives on contesting algorithmic decisions. Proc. of HCI 106:1--106:25

    work page 2021

  72. [72]

    Madumal, P.; Miller, T.; Sonenberg, L.; and Vetere, F. 2019. A grounded interaction protocol for explainable artificial intelligence. In Proc. of AAMAS , 1033--1041

    work page 2019

  73. [73]

    McBurney, P.; Parsons, S.; and Wooldridge, M. J. 2002. Desiderata for agent argumentation protocols. In Proc. of AAMAS , 402--409

    work page 2002

  74. [74]

    Miller, T. 2023. Explainable AI is dead, long live explainable ai!: Hypothesis-driven decision support using evaluative AI . In Proc. of FAccT , 333--342

    work page 2023

  75. [75]

    O.; Rotstein, N

    Moguillansky, M. O.; Rotstein, N. D.; Falappa, M. A.; Garc \' a, A. J.; and Simari, G. R. 2013. Dynamics of knowledge in DeLP through argument theory change. Theory Pract. Log. Program. 13(6):893--957

    work page 2013

  76. [76]

    Montavon, G.; Binder, A.; Lapuschkin, S.; Samek, W.; and M \"u ller, K.-R. 2019. Layer-wise relevance propagation: an overview. In Explainable AI: Interpreting, Explaining and Visualizing Deep Learning . Springer. 193--209

    work page 2019

  77. [77]

    Muhammad, K.; Ullah, A.; Lloret, J.; Del Ser, J.; and de Albuquerque, V. H. C. 2020. Deep learning for safe autonomous driving: Current challenges and future directions. IEEE Transactions on Intelligent Transportation Systems 22(7):4316--4336

    work page 2020

  78. [78]

    A.; Meel, K

    Narodytska, N.; Shrotri, A. A.; Meel, K. S.; Ignatiev, A.; and Marques - Silva, J. 2019. Assessing heuristic machine learning explanations with model counting. In Proc. of SAT , volume 11628, 267--278

    work page 2019

  79. [79]

    Niskanen, A. 2020. Computational Approaches to Dynamics and Uncertainty in Abstract Argumentation . Ph.D. Dissertation, University of Helsinki, Finland

    work page 2020

  80. [80]

    Oikarinen, E., and Woltran, S. 2011. Characterizing strong equivalence for argumentation frameworks. Artif. Intell. 175(14-15):1985--2009

    work page 2011

Showing first 80 references.