Många av frågorna framställda i diskussionen kräver ytterligare studier för att kunna besvaras.
En obesvarad fråga är hur skyddaren skall spela i spel där Stackelberg och Nash inte alls ger samma eller liknande resultat, ska vi använda Nash eller Stackelberg?
Intressant vore att testa med hjälp andra tillvägagångssätt, med fler algoritmer och försökspersoner än vad som användes i denna studie. Det skulle också vara till nytta att utföra test med mer beräkningskraft, under längre tid, eftersom det skulle öka möjligheten för algoritmen att förbättras ytterligare.
Om det är möjligt skulle det vara intressant att se resultaten av test utförda i en miljö utan begränsningarna, främst hos beräkningsförmågan samt ply-sökbarhet. För att komma till en mer fullständig förståelse av vilka faktorer som påverkar algoritmens förmåga är det
dessutom nödvändigt att undersöka fler parametrar.
Framtida utveckling hos botten borde inkludera utveckling och undersökning av Quiescence sök för att utvärdera Minimax-sök träden. Det finns också rum att lägga till flera datakällor för algoritmen, såsom öppningsdrag i schack kontext. Detta är något som sannolikt inte bara förbättrar anpassningsförmågan, men också flexibiliteten hos programmets bedömnings- och utvärderingsalgoritm. En ideal implementering av denna typ av AI skulle kunna analysera lära och analysera drag samt prestanda baserat på ett större dataset, med möjlighet att inte enbart förutspå och adaptera sin strategi 5–6 rundor framåt, utan djupare sök samt förmågan att adaptera till olika typer av spel, simulerade och verkliga scenarion.
Källor:
Hinton, G., Osindero, S. and Teh, Y. (2006). A Fast Learning Algorithm for Deep Belief Nets.
Neural Computation. Hämtad 21/12-18.
Hubel, D. and Wiesel, T. (1962). Receptive fields, binocular interaction and functional architecture in the cat's visual cortex. The Journal of Physiology, pp.106-154.
Silver, D., Hubert, T., Schrittwieser, J., Antonoglou, I., Lai, M., Guez, A., Lanctot, M., Sifre, L., Kumaran, D., Graepel, T., Lillicrap, T., Simonyan, K. and Hassabis, D. (2018). A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play.
Cornell University Science, p.10, pp. 4. Hämtad 21/12-18.
KingBase. (2018). KingBase - a free chess games database, updated monthly. [online] Tillgänglig via: http://www.kingbase-chess.net/ Hämtad 29/12-18.
Rabinowitz, P. (1986). Minimax methods in critical point theory with applications to
differential equations. Providence: American Mathematical Society, p. 8. Hämtad 29/12-18.
Reddy, D. (1977). Speech Understanding Systems: A Summary of Results of the Five-Year Research Effort. Department of Computer Science, p. 7-14. Hämtad 29/12-18.
Silver, D., Hubert, T., Schrittwieser, J., Antonoglou, I., Lai, M., Guez, A., Lanctot, M., Sifre, L., Kumaran, D., Graepel, T., Lillicrap, T., Simonyan, K. and Hassabis, D. (2018). A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play.
Science, 362(6419), pp.1140-1144.
Jacobsen, D. and Andersson, S. (2017). Hur genomför man undersökningar?. 2nd ed. Lund: Studentlitteratur AB, p.85.
Russell, S. and Norvig, P. (2016). Artificial intelligence. 3rd ed. New Jersey, p.174. Maschler, M., Solan, E., Hellman, Z., Borns, M. and Zamir, S. (2016). Game theory. Cambridge: Cambridge University Press, pp.176-180.
Shannon, C. (1950). XXII. Programming a computer for playing chess. The London, Edinburgh, And Dublin Philosophical Magazine And Journal Of Science, 41(314), 256-275. Shannon, C. (1950). XXII. Programming a computer for playing chess. The London, Edinburgh, And Dublin Philosophical Magazine And Journal Of Science, 41(314), 260. FICS Games Database - Download. (2019). Retrieved from
https://www.ficsgames.org/download.html
Benefits of Machine Learning for Your Business – Hacker Noon. (2019). Retrieved from https://hackernoon.com/benefits-of-machine-learning-for-your-business-624c7297a3af Heinz, E. A. (2000). A new self-play experiment in computer chess (No. 608). Technical Memo.
Elo, A. (2008). The rating of chessplayers, past and present (p. 18). Bronx, N.Y.: Ishi Press International.
Thrun, S. (1995). Learning to play the game of chess. In Advances in neural information
processing systems (pp. 1073).
Berliner, H. J. (1973, August). Some Necessary Conditions for a Master Chess Program. In
IJCAI (Vol. 3, pp. 77-85).
Beal, D. F. (1990). A generalised quiescence search algorithm. Artificial Intelligence, 43(1), 88-89.
python-chess. (2019). Retrieved from https://pypi.org/project/python-chess/
Streeter, W.F (1946). "Is the First Move an Advantage?". Chess Review. p. 16. Adorján 2004, “Black is Still OK!”, p. 68
Frey, C. B., & Osborne, M. A. (2017). The future of employment: how susceptible are jobs to computerisation?. Technological forecasting and social change, 114, 254-280.
U.S.-Canada Power System Outage Task Force, April 2004, U.S. Department of Energy. Causes of the August 14th blackout in the United States and Canada.
K. Chang (2003). New York Times. Ideal sensors for terror attack don’t exist yet. Israeli, E., & Wood, R. K. (2002). Shortest‐path network interdiction. Networks: An International Journal, 40(2), 97-111.
Policastro, A. J., & Gordon, S. P. (1999, May). The use of technology in preparing subway systems for chemical/biological terrorism. In Commuter Rail/Rapid Transit Conference Proceedings, Toronto.
Brown, G., Carlyle, M., Diehl, D., Kline, J., & Wood, K. (2005). A two-sided optimization for theater ballistic missile defense. Operations research, 53(5), 745-763.
Golden, B. (1978). A problem in network interdiction. Naval Research Logistics Quarterly, 25(4), 711-713.
Yin, X., Goudriaan, J. A. N., Lantinga, E. A., Vos, J. A. N., & Spiertz, H. J. (2003). A flexible sigmoid function of determinate growth. Annals of botany, 91(3), 361-371.
Brown, G. G., Carlyle, W. M., Salmeron, J., & Wood, K. (2005). Analyzing the vulnerability of critical infrastructure to attack and planning defenses. In Emerging Theory, Methods, and Applications (pp. 102-123). INFORMS.
Shieh, E., An, B., Yang, R., Tambe, M., Baldwin, C., DiRenzo, J., ... & Meyer, G. (2012, June). Protect: A deployed game theoretic system to protect the ports of the united states. In Proceedings of the 11th International Conference on Autonomous Agents and Multiagent Systems-Volume 1 (pp. 13-20). International Foundation for Autonomous Agents and Multiagent Systems.
Korzhyk, D., Conitzer, V., & Parr, R. (2011, May). Solving Stackelberg games with uncertain observability. In The 10th International Conference on Autonomous Agents and Multiagent
Systems-Volume 3 (pp. 1013-1020). International Foundation for Autonomous Agents and Multiagent Systems.
Jain, M., Kardes, E., Kiekintveld, C., Ordónez, F., & Tambe, M. (2010, July). Security games with arbitrary schedules: A branch and price approach. In Twenty-Fourth AAAI Conference on Artificial Intelligence.
Salmeron, J., Wood, K., & Baldick, R. (2004). Analysis of electric grid security under terrorist threat. IEEE Transactions on power systems, 19(2), 905-912.
Boin, A., & McConnell, A. (2007). Preparing for critical infrastructure breakdowns: the limits of crisis management and the need for resilience. Journal of Contingencies and Crisis
Management, 15(1), 50-59.
Critical Infrastructure Sectors, Homeland Security. (2019). Hämtad från https://www.dhs.gov/cisa/critical-infrastructure-sectors
Gibbs, M. N., & MacKay, D. J. (2000). Variational Gaussian process classifiers. IEEE
Transactions on Neural Networks, 11(6), 1458-1464.
Cybenko, G. (1989). Approximation by superpositions of a sigmoidal function. Mathematics
of control, signals and systems, 2(4), 303-314.
Hecht-Nielsen, R. (1992). Theory of the backpropagation neural network. In Neural networks
for perception (pp. 65-93). Academic Press.
Funahashi, K. I. (1989). On the approximate realization of continuous mappings by neural networks. Neural networks, 2(3), 183-192.
Han, J., & Moraga, C. (1995, June). The influence of the sigmoid function parameters on the speed of backpropagation learning. In International Workshop on Artificial Neural Networks (pp. 195-201). Springer, Berlin, Heidelberg.
WMATA (2016) Hämtad från
https://planitmetro.com/2016/03/24/data-download-metrorail-ridership-by-station-by-month-2010-2015/
PyPy (2007). Hämtad från
https://pypy.org/
UK/BM-56 TRANSLATION, FAS. (2004) Hämtad från
https://fas.org/irp/world/para/manualpart1_3.pdf
Berry, J. W., Fleischer, L., Hart, W. E., Phillips, C. A., & Watson, J. P. (2005). Sensor placement in municipal water networks. Journal of Water Resources Planning and Management, 131(3), 237-243.
Pita, J., Jain, M., Marecki, J., Ordóñez, F., Portway, C., Tambe, M., ... & Kraus, S. (2008, May). Deployed ARMOR protection: the application of a game theoretic model for security at the Los Angeles International Airport. In Proceedings of the 7th international joint conference on Autonomous agents and multiagent systems: industrial track (pp. 125-132). International Foundation for Autonomous Agents and Multiagent Systems.
Tsai, J., Kiekintveld, C., Ordonez, F., Tambe, M., & Rathi, S. (2009). IRIS-a tool for strategic security allocation in transportation networks.
Sion, M. (1958). On general minimax theorems. Pacific Journal of mathematics, 8(1), 171-176.
Nash, J. (1951). Non-cooperative games. Annals of mathematics, 286-295.
Binmore, K. (2007). Playing for real: a text on game theory. Oxford university press.