Yıl 2018, Cilt 6, Sayı 1, Sayfalar 93 - 107 2018-03-28

DEVELOPING A HUMAN CONTROLLED MODEL FOR SAFE ARTIFICIAL INTELLIGENCE SYSTEMS
GÜVENLİ YAPAY ZEKÂ SİSTEMLERİ İÇİN İNSAN DENETİMLİ BİR MODEL GELİŞTİRİLMESİ

Utku KÖSE [1]

48 60

Artificial Intelligence is known as one of the most effective research field of nowadays and the future. But rapid rise of Artificial Intelligence and its potential to solve all real world problems autonomously, it has caused also several anxieties. Some scientists think that intelligent systems can reach to a level, which is dangerous for the humankind so because of that some precautions should be taken. So, many sub-research fields like Machine Ethics or Artificial Intelligence Safety have appeared in time. In the context of the explanations so far, objective of this study is to suggest a secure intelligent model including human control and also precautions based on intelligent agent and Machine Learning. In the study, essential subjects regarding Artificial Intelligence Safety and details of the suggested model have been provided and also some evaluations about its potential have been done. It is though that this model can be an inspiration for safe Artificial Intelligence systems of the future.

Yapay Zekâ, gerek günümüz, gerekse geleceğin en etkin araştırma alanlarından birisi olarak bilinmektedir.  Ancak Yapay Zekâ’nın hızlı yükselişi ve otonom bir şekilde bütün gerçek dünya problemlerini çözebilir potansiyele sahip olması, çeşitli endişeleri de beraberinde getirmiştir. Bazı bilim insanları, zeki sistemlerin ilerleyen süreçte insanlığı tehdit edebilecek düzeye gelebileceğini ve bu nedenle çeşitli önlemlerin alınması gerektiğini düşünmektedir. Bu nedenle Makine Etiği ya da Yapay Zekâ Güvenliği gibi birçok alt-araştırma alanı da zaman içerisinde ortaya çıkmıştır. Açıklamalar bağlamında bu çalışmanın amacı da, insan denetimini de içeren, zeki etmen ve Makine Öğrenmesi odaklı önlemleri bünyesinde barındıran, güvenli bir zeki sistem modeli önermektir. Çalışmada Yapay Zekâ Güvenliği odaklı temel konularla birlikte önerilen modelin detaylarına ilişkin açıklamalar sunulmuş ve potansiyeli hakkında değerlendirmeler yapılmıştır. Modelin geleceğin güvenli Yapay Zekâ sistemlerine ilham kaynağı olabileceği düşünülmektedir.
  • Abbeel, P., Ng, A.Y., 2011. Inverse Reinforcement Learning. In Encyclopedia of Machine Learning (pp. 554-558). Springer US.
  • Alpaydın, E., 2014. Introduction to Machine Learning. MIT Press.
  • Amodei, D., Olah, C., Steinhardt, J., Christiano, P., Schulman, J., Mané, D., 2016. Concrete Problems in AI Safety. arXiv preprint arXiv:1606.06565.
  • Anderson, M., Anderson, S.L., 2007. Machine Ethics: Creating an ethical intelligent agent. AI Magazine, 28(4), 15.
  • Anderson, M., Anderson, S.L. (Eds.)., 2011. Machine Ethics. Cambridge University Press.
  • Armstrong, M.S., Orseau, L., 2016. Safely Interruptible Agents. Machine Intelligence Research Institute.
  • Arnold, T., Kasenberg, D., Scheutz, M., 2017. Value Alignment or Misalignment–What will Keep Systems Accountable. In 3rd International Workshop on AI, Ethics, and Society.
  • Ashrafian, H., 2015. Artificial intelligence and robot responsibilities: Innovating beyond rights. Science and Engineering Ethics, 21(2), 317-326.
  • Asimov, I., 2004. I, Robot (Vol. 1). ‘Güncel Bir Basım’. Spectra.
  • Awad, E., Dsouza, S., Rahwan, I., Shariff, A., Bonnefon, J.-F. 2018. Moral Machine. MIT Media Lab. Çevrimiçi (Erişim, 7 Şubat 2018): https://www.media.mit.edu/research/groups/10005/moral-machine
  • Barnett, D., 2017. The robots are coming – but will they really take all our jobs?. Independent – Web. Çevrimiçi (Erişim, 1 Şubat 2018): http://www.independent.co.uk/news/science/robots-are-coming-but-will-they-take-our-jobs-uk-artificial-intelligence-doctor-who-a8080501.html
  • Bostrom, N., 2002. Existential Risks. Journal of Evolution and Technology, 9(1), 1-31.
  • Bostrom, N., 2014. Superintelligence: Paths, Dangers, Strategies. OUP, Oxford.
  • Brady, R., 2017. The Doctor in the Machine: How AI Is Saving Lives in Healthcare. SingularityHub. Çevrimiçi (Erişim, 1 Şubat 2018): https://singularityhub.com/2017/11/30/the-doctor-in-the-machine-how-ai-is-saving-lives-in-healthcare/#sm.000077d60e4hhdn5t0g1x8tdyp8t4
  • Cellan-Jones, R., 2014. Hawking: Yapay zeka insanlığın sonunu getirebilir (Türkçe). BBC Türkçe – Web. Çevrimiçi (Erişim, 3 Şubat 2018): http://www.bbc.com/turkce/haberler/2014/12/141202_hawking_yapay_zeka
  • Cellan-Jones, R., 2017. The robot lawyers are here - and they’re winning. BBC News Technology – Web. Çevrimiçi (Erişim, 1 Şubat 2018): http://www.bbc.com/news/technology-41829534
  • Clarke, R., 1993. Asimov's Laws of Robotics: Implications for Information Technology-Part I. Computer, 26(12), 53-61.
  • Conitzer, V., Sinnott-Armstrong, W., Borg, J. S., Deng, Y., Kramer, M., 2017. Moral Decision Making Frameworks for Artificial Intelligence. In AAAI (pp. 4831-4835).
  • Copeland, J., 1993. Artificial Intelligence: A Philosophical Introduction, Blackwell: Oxford.
  • Dashevsky, E., 2017. Do Robots and AI Deserve Rights?. Enterpreneur – News and Trends – AI. Çevrimiçi (Erişim, 31 Ocak 2018): https://www.entrepreneur.com/article/289344
  • Davis, D., 2018. How AI and copyright would work. TechCrunch. Çevrimiçi (Erişim, 31 Ocak 2018): https://techcrunch.com/2018/01/09/how-ai-and-copyright-would-work/
  • Dewey, D., 2014. Reinforcement Learning and the Reward Engineering Principle. In 2014 AAAI Spring Symposium Series.
  • Dorigo, M., de Oca, M.A.M., Engelbrecht, A., 2008. Particle Swarm Optimization. Scholarpedia, 3(11), 1486.
  • Dormehl, L., 2017. I, Alexa: Should we give artificial intelligence human rights?. DigitalTrends – Computing. Çevrimiçi (Erişim, 31 Ocak 2018): https://www.digitaltrends.com/cool-tech/ai-personhood-ethics-questions/
  • Evans, O., Goodman, N.D., 2015. Learning the Preferences of Bounded Agents. In NIPS Workshop on Bounded Optimality.
  • Evans, O., Stuhlmüller, A., Goodman, N.D., 2016. Learning the Preferences of Ignorant, Inconsistent Agents. In AAAI (pp. 323-329).
  • Ferber, J., 1999. Multi-Agent Systems: An Introduction to Distributed Artificial Intelligence (Vol. 1). Reading: Addison-Wesley.
  • Galeon, D., Houser, K., 2017. Google’s Artificial Intelligence Built an AI That Outperforms Any Made by Humans, Futurism. Çevrimiçi (Erişim, 4 Şubat 2018): https://futurism.com/google-artificial-intelligence-built-ai/
  • Goodfellow, I., Bengio, Y., Courville, A., Bengio, Y., 2016. Deep Learning (Vol. 1). MIT Press.
  • Goodfellow, I., Papernot, N., Huang, S., Duan, Y., Abbeel, P., Clark, J., 2017. Attacking Machine Learning with Adversarial Examples, Open AI – Blog Web. Çevrimiçi (Erişim 5 Şubat 2018): https://blog.openai.com/adversarial-example-research/
  • Gubbi, J., Buyya, R., Marusic, S., Palaniswami, M., 2013. Internet of Things (IoT): A Vision, Architectural Elements, and Future Directions. Future Generation Computer Systems, 29(7), 1645-1660.
  • Hady, M.F.A., Schwenker, F., 2013. Semi-supervised Learning. In Handbook on Neural Information Processing (pp. 215-239). Springer Berlin Heidelberg.
  • Heath, N., 2015. Why AI could destroy more jobs than it creates, and how to save them. TechRepublic.com. Çevrimiçi (Erişim, 1 Şubat 2018): https://www.techrepublic.com/article/ai-is-destroying-more-jobs-than-it-creates-what-it-means-and-how-we-can-stop-it/
  • Holland, O. (Ed.)., 2003. Machine Consciousness. Imprint Academic.
  • John Walker, S., 2014. Big Data: A Revolution That Will Transform How We Live, Work, and Think, International Journal of Advertising, 33(1), 181-183.
  • Karaboğa, D., 2014. Yapay Zeka Optimizasyon Algoritmaları. Nobel Yayıncılık.
  • Kober, J., Peters, J., 2012. Reinforcement Learning in Robotics: A Survey. In Reinforcement Learning (pp. 579-610). Springer Berlin Heidelberg.
  • Kose, U., Pavaloiu, A., 2017. Dealing with Machine Ethics in Daily Life: A View with Examples. The 5th International Virtual Conference on Advanced Scientific Results. Slovakia, pp. 200-205. 10.18638/scieconf.2017.5.1.454.
  • Kotsiantis, S.B., 2007. Supervised Machine Learning: A Review of Classification Techniques. Informatica, 31, 249-268.
  • Kulaklı, G., 2017. Yüzyılın Kavgası: Mark Zuckerberg İle Elon Musk Birbirine Girdi!. WebTekno. Çevrimiçi (Erişim, 3 Şubat 2018): http://www.webtekno.com/yuzyilin-kavgasi-mark-zuckerberg-ile-elon-musk-birbirine-girdi-h31650.html
  • Kurzweil, R., 2005. The Singularity is Near: When Humans Transcend Biology. Penguin.
  • LeCun, Y., Bengio, Y., Hinton, G., 2015. Deep Learning. Nature, 521(7553), 436.
  • Maes, P. (Ed.)., 1990. Designing Autonomous Agents: Theory and Practice From Biology to Engineering and Back. MIT Press.
  • Massachusetts Teknoloji Enstitüsü, 2018. Moral Machine. Moral Machine Web. Çevrimiçi (Erişim, 7 Şubat 2018): http://moralmachine.mit.edu/
  • Metz, C., 2017. Building A.I. That Can Build A.I., The New York Times. Çevrimiçi (Erişim, 4 Şubat 2018): https://www.nytimes.com/2017/11/05/technology/machine-learning-artificial-intelligence-ai.html
  • Minsky, M., 2007. The Emotion Machine: Commonsense Thinking, Artificial Intelligence, and the Future of the Human Mind. Simon and Schuster.
  • Mnih, V., Kavukcuoglu, K., Silver, D., Graves, A., Antonoglou, I., Wierstra, D., Riedmiller, M., 2013. Playing Atari with Deep Reinforcement Learning. arXiv preprint arXiv:1312.5602.
  • Moor, J., 2009. Four Kinds of Ethical Robots. Philosophy Now, 72, 12-14.
  • Muehlhauser, L., Helm, L., 2012. The Singularity and Machine Ethics. In Singularity Hypotheses (pp. 101-126). Springer, Berlin, Heidelberg.
  • Murphy, R., Woods, D.D., 2009. Beyond Asimov: The Three Laws of Responsible Robotics. IEEE Intelligent Systems, 24(4).
  • Nabiyev, V.V., 2005. Yapay Zeka: Problemler-Yöntemler-Algoritmalar. Seçkin Yayıncılık.
  • Ng, A.Y., Russell, S.J., 2000. Algorithms for Inverse Reinforcement Learning. In ICML (pp. 663-670).
  • Norman, A., 2018. Your Future Doctor May Not be Human. This Is the Rise of AI in Medicine.. Futurism – SciFi Visions. Çevrimiçi (Erişim, 1 Şubat 2018): https://futurism.com/ai-medicine-doctor/
  • Orseau, L., Armstrong, S., 2016. Safely Interruptible Agents. In Uncertainty in Artificial Intelligence: 32nd Conference (UAI 2016), (Eds.) Alexander Ihler and Dominik Janzing, (pp. 557-566).
  • Pavaloiu, A., Kose, U., 2017. Ethical Artificial Intelligence-An Open Question. Journal of Multidisciplinary Developments, 2(2), 15-27.
  • Riedl, M.O., Harrison, B., 2016. Using Stories to Teach Human Values to Artificial Agents. In AAAI Workshop: AI, Ethics, and Society.
  • Russell, S.J., Norvig, P., Canny, J.F., Malik, J.M., Edwards, D.D., 2003. Artificial Intelligence: A Modern Approach (Vol. 2, No. 9). Upper Saddle River: Prentice Hall.
  • Russell, S., Dewey, D., Tegmark, M., 2015. Research Priorities for Robust and Beneficial Artificial Intelligence. Ai Magazine, 36(4), 105-114.
  • Schneider, S., 2016. Science Fiction and Philosophy: From Time Travel to Superintelligence. John Wiley & Sons.
  • Shead, S., 2016. Google has developed a 'big red button' that can be used to interrupt artificial intelligence and stop it from causing harm, Business Insider UK. Çevrimiçi (Erişim, 5 Şubat 2018): http://uk.businessinsider.com/google-deepmind-develops-a-big-red-button-to-stop-dangerous-ais-causing-harm-2016-6
  • Silva, T.C., Zhao, L., 2016. Network-Based Unsupervised Learning. In Machine Learning in Complex Networks (pp. 143-180). Springer International Publishing.
  • Singh, S., 2018. Will Artificial Intelligence take over jobs?. The Economic Times (India Times) – Web. Çevrimiçi (Erişim, 1 Şubat 2018): https://economictimes.indiatimes.com/tech/ites/will-artificial-intelligence-take-over-jobs/articleshow/62610145.cms
  • Sutton, R.S., Barto, A.G., 1998. Reinforcement Learning: An Introduction (Vol. 1, No. 1). MIT Press.
  • The Associated Press, 2017. For Driverless Cars, a Moral Dilemma: Who Lives and Who Dies?, NBC News Web. Çevrimiçi (Erişim, 7 Şubat 2018): http://www.nbcnews.com/tech/innovation/driverless-carsmoral-dilemma-who-lives-who-dies-n708276
  • The Week, 2018. Amazon Go: AI-powered supermarket opens. The Week – Artificial Intelligence. Çevrimiçi (Erişim, 1 Şubat 2018): http://www.theweek.co.uk/artificial-intelligence/91111/amazon-go-ai-powered-supermarket-opens
  • Vamplew, P., Dazeley, R., Foale, C., Firmin, S., Mummery, J., 2017. Human-Aligned Artificial Intelligence is a Multiobjective Problem. Ethics and Information Technology, 1-14.
  • Wu, X., Zhu, X., Wu, G.Q., Ding, W., 2014. Data Mining with Big Data. IEEE Transactions on Knowledge and Data Engineering, 26(1), 97-107.Yampolskiy, R.V., 2013. Artificial Intelligence Safety Engineering: Why Machine Ethics is a Wrong Approach. In Philosophy and theory of artificial intelligence (pp. 389-396). Springer, Berlin, Heidelberg.
  • Yampolskiy, R.V., 2015. Artificial Superintelligence: A Futuristic Approach. CRC Press.
  • Xia, F., Yang, L.T., Wang, L., Vinel, A., 2012. Internet of Things. International Journal of Communication Systems, 25(9), 1101.
Birincil Dil tr
Konular Mühendislik ve Temel Bilimler
Dergi Bölümü Araştırma Makaleleri \ Research Articles
Yazarlar

Orcid: 0000-0002-9652-6415
Yazar: Utku KÖSE (Sorumlu Yazar)
E-posta: utkukose@sdu.edu.tr
Kurum: Süleyman Demirel Üniversitesi
Ülke: Turkey


Bibtex @araştırma makalesi { jesd394527, journal = {Mühendislik Bilimleri ve Tasarım Dergisi}, issn = {}, address = {Süleyman Demirel Üniversitesi}, year = {2018}, volume = {6}, pages = {93 - 107}, doi = {10.21923/jesd.394527}, title = {GÜVENLİ YAPAY ZEKÂ SİSTEMLERİ İÇİN İNSAN DENETİMLİ BİR MODEL GELİŞTİRİLMESİ}, key = {cite}, author = {KÖSE, Utku} }
APA KÖSE, U . (2018). GÜVENLİ YAPAY ZEKÂ SİSTEMLERİ İÇİN İNSAN DENETİMLİ BİR MODEL GELİŞTİRİLMESİ. Mühendislik Bilimleri ve Tasarım Dergisi, 6 (1), 93-107. DOI: 10.21923/jesd.394527
MLA KÖSE, U . "GÜVENLİ YAPAY ZEKÂ SİSTEMLERİ İÇİN İNSAN DENETİMLİ BİR MODEL GELİŞTİRİLMESİ". Mühendislik Bilimleri ve Tasarım Dergisi 6 (2018): 93-107 <http://dergipark.gov.tr/jesd/issue/33308/394527>
Chicago KÖSE, U . "GÜVENLİ YAPAY ZEKÂ SİSTEMLERİ İÇİN İNSAN DENETİMLİ BİR MODEL GELİŞTİRİLMESİ". Mühendislik Bilimleri ve Tasarım Dergisi 6 (2018): 93-107
RIS TY - JOUR T1 - GÜVENLİ YAPAY ZEKÂ SİSTEMLERİ İÇİN İNSAN DENETİMLİ BİR MODEL GELİŞTİRİLMESİ AU - Utku KÖSE Y1 - 2018 PY - 2018 N1 - doi: 10.21923/jesd.394527 DO - 10.21923/jesd.394527 T2 - Mühendislik Bilimleri ve Tasarım Dergisi JF - Journal JO - JOR SP - 93 EP - 107 VL - 6 IS - 1 SN - -1308-6693 M3 - doi: 10.21923/jesd.394527 UR - http://dx.doi.org/10.21923/jesd.394527 Y2 - 2018 ER -
EndNote %0 Mühendislik Bilimleri ve Tasarım Dergisi GÜVENLİ YAPAY ZEKÂ SİSTEMLERİ İÇİN İNSAN DENETİMLİ BİR MODEL GELİŞTİRİLMESİ %A Utku KÖSE %T GÜVENLİ YAPAY ZEKÂ SİSTEMLERİ İÇİN İNSAN DENETİMLİ BİR MODEL GELİŞTİRİLMESİ %D 2018 %J Mühendislik Bilimleri ve Tasarım Dergisi %P -1308-6693 %V 6 %N 1 %R doi: 10.21923/jesd.394527 %U 10.21923/jesd.394527