Abstract
There is a core problem with the modern Artificial Intelligence (AI) technologies, based on the current new wave of Artificial Neural Networks (ANNs). Whether they have been used in healthcare or for exploring Mars, we, the programmers who build them, do not know well why they make some decisions over others. Many are putting into question, hence, this aura of AI objectivity and infallibility; on our side, we, instead, identify a key issue around the problem of AI errors and bias into an insufficient human ability to determine the limits of the context, where the ANNs will have to operate. In fact, while it is of great amplitude the range of what the rational side of the human mind can master, machine intelligence has limited capacity to learn in completely unknown scenarios. Simply, an inaccurate or incomplete codification of the context may result into AI failures. We present here a simple cognification ANN-based case study, in an underwater scenario, where the difficulty of identifying and then codifying all the relevant contextual features has led to a situation of partial failure. This paper reports on our reflections, and subsequent technical actions taken to recover from this situation.
Similar content being viewed by others
References
Lemley J, Bazrafkan S, Corcoran P (2017) Deep learning for consumer devices and services: pushing the limits for machine learning, artificial intelligence, and computer vision. IEEE Consum Electron Mag 6.2:48–56
Wu Y, Schuster M, Chen Z, Le QV, Norouzi M, Macherey W, Krikun M, Cao Y, Gao Q, Macherey K, Klingner J, Shah A, Johnson M, Liu X, Kaiser L, Gouws S, Kato Y, Kudo T, Kazawa H, Stevens K, Kurian G, Patil N, Wang W, Young C, Smith J, Riesa J, Rudnick A, Vinyals O, Corrado G, Hughes M, Dean J (2016) Google’s neural machine translation system: bridging the gap between human and machine translation. CoRR abs/1609.08144 (2016). arXiv:1609.08144 http://arxiv.org/abs/1609.08144
Van Den Oord A, Dieleman S, Zen H, Simonyi K, Vinyals O, Graves A, Kalchbrenner N, Senior AW, Kavukcuoglu K (2016) WaveNet: A generative model for raw audio. In SSW 125
Son Chung J, Zisserman A (2018) Learning to lip read words by watching videos. Comput Vis Image Underst
S. Suwajanakorn, S. M. Seitz, and I. Kemelmacher-Shlizerman. 2017. Synthesizing Obama: learning lip sync from audio. ACM Trans. Graph. 36, 4, Article 95 (July 2017), 13 pages. doi:https://doi.org/10.1145/3072959.3073640
Chen F, Li B, Dong R, Zhao P (2018) High-performance OCR on packing boxes in industry based on deep learning. Pacific Rim International Conference on Artificial Intelligence Springer, 1018–1030
Silver D, Huang A, Maddison CJ, Guez A, Sifre L, Van Den Driessche G, Schrittwieser J, Antonoglou I, Panneershelvam V, Lanctot M et al (2016) Mastering the game of go with deep neural networks and tree search. Nature 529 7587(2016):484
Moravˇc’ık M, Schmid M, Burch N, Lis’y V, Morrill D, Bard N et al (2017) Deepstack: expert-level artificial intelligence in heads-up no-limit poker. Science 356(6337):508–513
AI O, (2018) The open ai dota 2 bot. [Accessed 20-August-2018]. [Online]. Available: https://blog.openai.com/more-on-dota-2/
Ferretti S, Mirri S, Prandi C, Salomoni P (2016) Automatic web content personalization through reinforcement learning. J Syst Softw 121:157–169. https://doi.org/10.1016/j.jss.2016.02.008
Roccetti M, Salomoni P, Prandi C, Marfia G, Montagnani M, Gningaye L (2016) Understanding Crohn’s disease patients reaction to infliximab from facebook: A medical perspective. Proceedings of the 2016 IEEE/ACM international conference on advances in social networks analysis and mining, ASONAM 2016. 1007 – 1010. doi:https://doi.org/10.1109/ASONAM.2016.7752364
Roccetti M, Prandi C, Salomoni P, Marfia G (2016) Unleashing the true potential of social networks: confirming infliximab medical trials through Facebook posts. Netw Modeling Anal Health Inform Bioinform 5:1. https://doi.org/10.1007/s13721-016-0122-9
Silvia Mirri, Catia Prandi, Marco Roccetti, and Paola Salomoni. (2017) Handmade narrations: handling digital narrations on food and gastronomic culture. ACM J Comput Cult Herit. doi:https://doi.org/10.1145/3097569
Ferretti S, Mirri S, Prandi C, Salomoni P (2017) On personalizing web content through reinforcement learning. Univ Access Inf Soc 2017. https://doi.org/10.1007/s10209-016-0463-2
Melis A, Mirri S, Prandi C, Prandini M, Salomoni P, Callegati F (2018) Integrating personalized and accessible itineraries in MaaS ecosystems through microservices. Mobile Networks and Applications 2018. https://doi.org/10.1007/s11036-017-0831-z
Caruana R, Lou Y, Gehrke J, Koch P, Sturm M, Elhadad N (2015) Intelligible models for healthcare: predicting pneumonia risk and hospital 30-day readmission. Proceedings of the 21th ACM SIGKDD international conference on knowledge discovery and data mining. ACM: 1721–1730
Melis A, Mirri S, Prandi C, Prandini M, Salomoni P, Callegati F (2016) Crowd sensing for smart mobility through a service-oriented architecture. IEEE 2nd Int Smart Cities Conf: Improving Citizens Qual Life, ISC2 2016. https://doi.org/10.1109/ISC2.2016.7580861
Mirri S, Prandi C, Salomoni P, Monti L (2016) Social location awareness: A prototype of altruistic IoT. 2016 8th IFIP International Conference on New Technologies, Mobility and Security, NTMS 2016. https://doi.org/10.1109/NTMS.2016.7792477
Zou J, Schiebinger L (2018) AI can be sexist and racist – it’s time to make it fair. Nature 559:324–326
Buolamwini J, Gebru T (2018) Gender shades: intersectional accuracy disparities in commercial gender classification. Conf Fairness Account Transparency 77–91
Delnevo G, Lipparini A, Roccetti M, Sobrero M (2018) Human intelligence, ambidexterity and ai: reflections on a data cognification case study in an underwater scenario. 2nd annual science fiction prototyping conference 2018: designing your future with science fiction. SciFi-It 2018(2018):50–55
Delnevo G, Roccetti M, Mirri S (2018) Intelligent machines for good?: more focus on the context. Goodtechs '18 international conference on smart objects and Technologies for Social Good Bologna, Italy
Graham R, McCabe H, Sheridan S (2004) Neural networks for real-time pathfinding in computer games. ITB J 5(1):21
Kindermann T, Cruse H, Dautenhahn K (1996) A fast, three-layer neural network for path finding. Netw Comput Neural Syst 7(2):423–436
Kim H, Son H, Roska T, Chua LO (2002) Optimal path finding with space- and time-variant metric weights via multi-layer CNN. Int J Circuit Theory Applic 30(2–3):247–270
Benardos PG, Vosniakos G-C (2007) Optimizing feedforward artificial neural network architecture. Eng Appl Artif Intell 20.3:365–382
Fails JA, Olsen DR Jr. (2003) Interactive machine learning. Proc 8th Int Conf Intell User Interf. ACM
Arino A, De La Torre J (1998) Learning from failure: towards an evolutionary model of collaborative ventures. Organ Sci 9(3):306–325
Coelho PRP, McClure JE (2005) Learning from failure. Am J Business 20(1):1
Shepherd DA (2003) Learning from business failure: propositions of grief recovery for the self-employed. Acad Manag Rev 28(2):318–328
Pan SJ, Yang O (2010) A survey on transfer learning. IEEE Trans Knowl Data Eng 22.10:1345–1359
Acknowledgements
We are grateful to the company that has provided us with the experimental seabeds. Topologies and company name are kept secret for the sake of privacy and security. Our thanks also go to Maurizio Sobrero, Andrea Lipparini, Lorenzo Donatiello and Gustavo Marfia (University of Bologna) for many precious suggestions on a previous version of this manuscript.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Delnevo, G., Roccetti, M. & Mirri, S. Intelligent and Good Machines? The Role of Domain and Context Codification. Mobile Netw Appl 25, 977–985 (2020). https://doi.org/10.1007/s11036-019-01233-7
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11036-019-01233-7