No Access

Metaverse: A Solution to the Multi-Agent Value Alignment Problem

Computer Science and Engineering, University of Louisville, 222 Eastern Parkway, Louisville, KY 40292, USA

E-mail Address: roman.yampolskiy@louisville.edu

https://doi.org/10.1142/S2705078522500072Cited by:4 (Source: Crossref)

Abstract

AI Safety researchers attempting to align values of highly capable intelligent systems with those of humanity face a number of challenges including personal value extraction, multi-agent value merger and finally in-silico encoding. State-of-the-art research in value alignment shows difficulties in every stage in this process, but merger of incompatible preferences is a particularly difficult challenge to overcome. In this paper, we assume that the value extraction problem will be solved and propose a possible way to implement an AI solution which optimally aligns with individual preferences of each user. We conclude by analyzing benefits and limitations of the proposed approach.

Keywords:

References

Abbeel, P. and Ng, A. Y. [2004] “Apprenticeship learning via inverse reinforcement learning,” in Paper Presented at the Proc. Twenty-First Int. Conf. Machine Learning. Crossref, Google Scholar
Anonymous. [2006]. Wish for immortality 1.1, in Paper Presented at the The Open-Source Wish Project, Available at: http://www.homeonthestrange.com/phpBB2/viewforum.php?f=4. Google Scholar
Armstrong, S., Sandberg, A. and Bostrom, N. [2012] Thinking inside the box: Controlling and using an oracle AI, Minds Mach. 22(4), 299–324. Crossref, Google Scholar
Arrow, K. J. [1950] A difficulty in the concept of social welfare, J. Polit. Econ. 58(4), 328–346. Crossref, Google Scholar
Arrow, K. J. [2012] Social Choice and Individual Values (Vol. 12) (Yale University Press). Google Scholar
Babcock, J., Kramár, J. and Yampolskiy, R. [2016], The AGI containment problem, in Artificial General Intelligence (Springer), pp. 53–63. Crossref, Google Scholar
Babcock, J., Kramar, J. and Yampolskiy, R. V. [2017] Guidelines for Artificial Intelligence Containment. arXiv preprint arXiv:1707.08476. Google Scholar
Baudrillard, J. [1994] Simulacra and Simulation (University of Michigan Press). Google Scholar
Bostrom, N. [2003a] Are we living in a computer simulation? Philos. Quar. 53(211), 243–255. Crossref, Google Scholar
Bostrom, N. [2003b] Astronomical waste: The opportunity cost of delayed technological development, Utilitas 15(3), 308–314. Crossref, Google Scholar
Bostrom, N. [2009] Moral uncertainty – towards a solution? in Paper presented at the Overcoming Bias, Available at: at: http://www.overcomingbias.com/2009/01/moral-uncertainty-towards-a-solution.html. Google Scholar
Bostrom, N. [2014] Superintelligence: Paths, Dangers, Strategies (Oxford University Press). Google Scholar
Boulos, M. N. K., Hetherington, L. and Wheeler, S. [2007] Second Life: an overview of the potential of 3-D virtual worlds in medical and health education, Health Inform. Lib. J. 24(4), 233–245. Crossref, Google Scholar
Brundage, M., Avin, S., Clark, J., Toner, H., Eckersley, P., Garfinkel, B., $\dots$ Filar, B. [2018] The malicious use of artificial intelligence: Forecasting, prevention, and mitigation. arXiv preprint arXiv:1802.07228. Google Scholar
Bushell, W. C. and Seaberg, M. [December 5, 2018] Experiments suggest humans can directly observe the quantum, in Paper Presented at the Psychology Today, Available at: https://www.psychologytoday.com/us/blog/sensorium/201812/experiments-suggest-humans-can-directly-observe-the-quantum. Google Scholar
Carr, B. [2007] Universe or Multiverse? (Cambridge University Press). Crossref, Google Scholar
Chalmers, D. J. [1996] The Conscious Mind: In Search of a Fundamental Theory (Oxford University Press). Google Scholar
Chalmers, D. [2003], The Matrix as metaphysics, in Science Fiction and Philosophy From Time Travel to Superintelligence, 36. Google Scholar
Chalmers, D. J. [2017] The virtual and the real, Disputatio 9(46), 309–352. Crossref, Google Scholar
Dignum, V. [2017] Responsible artificial intelligence: Designing AI for human values, ITU J ICT Discov. Google Scholar
Eco, U. [1990] Travels in Hyper Reality: Essays (Houghton Mifflin Harcourt). Google Scholar
Evans, O., Stuhlmüller, A. and Goodman, N. D. [2016] Learning the preferences of ignorant, inconsistent agents, in Paper presented at the AAAI. Google Scholar
Everitt, T., Lea, G. and Hutter, M. [2018] AGI Safety Literature Review. arXiv preprint arXiv:1805.01109. Google Scholar
Faggella, D. [August 17, 2018] Substrate Monopoly – The Future of Power in a Virtual and Intelligent World. Available at: https://danfaggella.com/substrate-monopoly/. Google Scholar
Faggella, D. [August 27, 2018] Programmatically Generated Everything (PGE). Available at: https://danfaggella.com/programmatically-generated-everything-pge/. Google Scholar
Faggella, D. [July 15, 2018] Digitized and Digested. Available at: https://danfaggella.com/digitized-and-digested/. Google Scholar
Faggella, D. [May 14, 2013] Transhuman Possibilities and the “Epitome of Freedom”. Available at: https://danfaggella.com/transhuman-possibilities-and-the-epitome-of-freedom/. Google Scholar
Faggella, D. [May 28, 2018] The Transhuman Transition – Lotus Eaters vs World Eaters. Available at: https://danfaggella.com/the-transhuman-transition-lotus-eaters-vs-world-eaters/. Google Scholar
Feygin, Y. B., Morris, K. and Yampolskiy, R. V. [2018] Uploading Brain into Computer: Whom to Upload First? arXiv preprint arXiv:1811.03009. Google Scholar
Feynman, R. and Gilbert, D. [1961] Miniaturization (Reinhold, New York), pp. 282–296. Google Scholar
Fuller, R. B. [1982] Synergetics: Explorations in the Geometry of Thinking (Estate of R. Buckminster Fuller). Google Scholar
Gardner, J. N. [2003] Biocosm: The New Scientific Theory of Evolution: Intelligent Life is the Architect of the Universe (Inner Ocean Publishing). Google Scholar
Gehrlein, W. V. [2002] Condorcet’s paradox and the likelihood of its occurrence: different perspectives on balanced preferences, Theory Decis. 52(2), 171–199. Crossref, Google Scholar
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., $\dots$ Bengio, Y. [2014] Generative adversarial nets, in Paper presented at the Advances in neural information processing systems. Google Scholar
Hanson, R. [2016] The Age of Em: Work, Love, and Life when Robots Rule the Earth (Oxford University Press). Crossref, Google Scholar
Harnad, S. [1992] The Turing Test is not a trick: Turing indistinguishability is a scientific criterion, ACM SIGART Bull. 3(4), 9–10. Crossref, Google Scholar
Harvey, D. [1989] The Condition of Postmodernity (Vol. 14) (Blackwell, Oxford). Google Scholar
High, R. [2012] The Era of Cognitive Systems: An Inside Look at IBM Watson and How it Works (IBM Corporation, Redbooks). Google Scholar
Kim, T. W., Donaldson, T. and Hooker, J. [2018] Mimetic vs Anchored Value Alignment in Artificial Intelligence. arXiv preprint arXiv:1810.11116. Google Scholar
Knight, W. [December 3, 2018] AI software can dream up an entire digital world from a simple sketch. Available at: https://www.technologyreview.com/s/612503/ai-software-can-dream-up-an-entire-digital-world-from-a-simple-sketch. Google Scholar
Krizhevsky, A., Sutskever, I. and Hinton, G. E. [2012] Imagenet classification with deep convolutional neural networks, in Paper presented at the Advances in Neural Information Processing Systems. Google Scholar
Kurzweil, R. and Grossman, T. [2005] Fantastic Voyage: Live Long Enough to Live Forever (Rodale). Google Scholar
Lévy, P. and Bono nno, R. [1998] Becoming Virtual: Reality in the Digital Age (Da Capo Press, Incorporated). Google Scholar
Lebens, S. and Goldschmidt, T. [2017] The Promise of a New Past (Michigan Publishing, University of Michigan Library, Ann Arbor, MI). Google Scholar
Lehman, J., Clune, J. and Misevic, D. [2018] The surprising creativity of digital evolution, in Paper presented at the Artificial Life Conference Proceedings. Google Scholar
Leskovec, J., Kleinb erg, J. and Faloutsos, C. [2005] Graphs over time: densification laws, shrinking diameters and possible explanations, in Paper presented at the Proc. eleventh ACM SIGKDD Int. Conf. Knowledge discovery in data mining. Google Scholar
Lowrance, C. J., A bdelwahab, O. and Yampolskiy, R. V. [2015] Evolution of a metaheuristic for aggregating wisdom from artificial crowds, in Paper presented at the Portuguese Conf. on Artificial Inte lligence. Google Scholar
Manheim, D. and Garrabrant, S. [2018] Categorizing Variants of Goodhart’s Law. arXiv preprint arXiv:1803. 04585. Google Scholar
Mnih, V., Kavuk cuoglu, K., Silver, D., Rusu, A. A., Veness, J., Bellemare, M. G., … Ostrovski, G. [2015] Hu man-level control through deep reinforcement learning, Nature 518(7540), 529–533. Crossref, Google Scholar
Moravčík, M., Schmid, M., Burch, N., Lisý, V., Morrill, D., Bard, N., … Bowling, M. [2017] Deepstack: Expert-level artificial intelligence in heads-up no-limit poker, Science 356(6337), 508–513. Crossref, Google Scholar
Muehlhauser, L. and Williamson, C. [2013] Ideal Advisor Theories and Personal CEV (Machine Intelligence Research Institute). Google Scholar
Mulgan, T. [2008] Future people: A moderate consequentialist account of our obligations to future generations. OUP Catalogue. Google Scholar
Negroponte, N., Harrington, R., McKay, S. R. and Christian, W. [1997] Being digital, Comput. Phys. 11(3), 261–262. Crossref, Google Scholar
Ng, A. Y. and Russell, S. J. [2000] Algorithms for inverse reinforcement learning, Paper presented at the Icm l. Google Scholar
Parfit, D. [1984] Reasons and Persons (OUP, Oxford). Google Scholar
Pearce, D. [1995] Hedonistic Imperative (David, Pearce). Google Scholar
Pistono, F. and Yampolskiy, R. V. [2016] Unethical Research: How to Create a Malevolent Artificial Intell igence. arXiv preprint arXiv:1605.02817. Google Scholar
Purves, D., Jenkins, R. and Strawser, B. J. [2015] Autonomous machines, moral judgment, and acting for the right reasons, Ethic. Theory Moral Prac. 18(4), 851–872. Crossref, Google Scholar
Putnam, H. [1981] Brain in a Vat. Reason, Truth History, 1–21. Google Scholar
Rand, A. [1964] The Virtue of Selfishness (Penguin). Google Scholar
Rand, A. [1988] The Ayn Rand Lexicon: Objectivism from A to Z (Vol. 4) (Penguin). Google Scholar
Raou lt, A. and Yampolskiy, R. [2015] Reviewing Tests for Machine Consciousness. Available at https://www.researchgate.net/publication/284859013_DRAFT_Reviewing_Tests_for_ Machine_Consciousness. Google Scholar
Ried l, M. O. and Harrison, B. [2016] Using stories to teach human values to artificial agents, in Paper presented at the AAAI Workshop: AI, Ethics, and Society. Google Scholar
Sarma, G. P., Hay, N. J. and Safron, A. [2018] AI safety and reproducibility: establishing robust foundations for the neuropsychology of human values, in Paper presented at the International Conference on Computer Safety, Reliability, and Security. Google Scholar
Schweizer, P. [1998] The truly total Turing test, Minds Mach. 8(2), 263–272. Crossref, Google Scholar
Silver, D., Schrittwieser, J., Simonyan, K., Antonoglou, I., Huang, A., Guez, A., … Bolton, A. [2017] Mastering the game of Go without human knowledge, Nature 550(7676), 354. Crossref, Google Scholar
Silver, D., Hubert, T., Schrittwieser, J., Antonoglou, I., Lai, M., Guez, A., … Graepel, T. [2018] A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play, Science 362(6419), 1140–1144. Crossref, Google Scholar
Simpson, Z. B. [2000] The in-game economics of Ultima Online, in Paper presented at the Computer Game Developer’s Conference, San Jose, CA. Google Scholar
Smart, J. M. [2012] The transcension hypothesis: Sufficiently advanced civilizations invariably leave our universe, and implications for METI and SETI, Acta Astronautica 78, 55–68. Crossref, Google Scholar
Soares, N. and Fallenstein, B. [2014] Aligning superintelligence with human interests: A technical research agenda, Machine Intelligence Research Institute (MIRI) technical report, 8. Google Scholar
Sobel, D. [1994] Full information accounts of well-being, Ethics 104(4), 784–810. Crossref, Google Scholar
Sotala, K. and Yampolskiy, R. V. [2014] Responses to catastrophic AGI risk: A survey, Phys. Scrip. 90(1), 018001. Crossref, Google Scholar
Tegmark, M. [2017] Life 3.0: Being Human in the Age of Artificial Intelligence (Knopf). Google Scholar
Trazzi, M. and Yampolskiy, R. V. [2018] Building safer AGI by introducing artificial stupidity. arXiv preprint arXiv:1808.03644. Google Scholar
Turchin, A. [2018] Wireheading as a Possible Contributor to Civilizational Decline. Available at: https://philpapers.org/rec/TURWAA. Google Scholar
Vallentyne, P. [2014], Robert Nozick: Anarchy, State, and Utopia, in Central Works of Philosophy v5 (Routledge), pp. 108–125. Google Scholar
Vilenkin, A. and Tegmark, M. [2011] The case for parallel universes. Scientific American. Retrieved from: http://www.scientificamerican.com/article/multiverse-the-case-for-parallel-universe. Google Scholar
Visions of a Better World. [December 19, 2018] Paper presented at the Scientific American, Available at: https://blogs.scientificamerican.com/cross-check/visions-of-a-better-world. Google Scholar
Warren, M. A. [1977] Do potential people have moral rights? Can. J. Philos. 7(2), 275–289. Crossref, Google Scholar
Wernick, I. K., Herman, R., Govind, S. and Ausubel, J. H. [1996] Materialization and dematerialization: measures and trends, Daedalus 171–198. Google Scholar
Yampolskiy, R. V., Ashby, L. and Hassan, L. [2012] Wisdom of artificial crowds—a metaheuristic algorithm for optimization, J. Intell. Learn. Syst. Appl. 4(2), 98–107. Google Scholar
Yampolskiy, R. V., Klare, B. and Jain, A. K. [2012] Face recognition in the virtual world: Recognizing Avatar faces, in Paper presented at the 2012 11th International Conference on Machine Learning and Applications (ICMLA). Crossref, Google Scholar
Yampolskiy, R. V. and Gavrilova, M. L. [2012] Artimetrics: biometrics for artificial entities, IEEE Robot. Automat. Magaz. 19(4), 48–58. Crossref, Google Scholar
Yampolskiy, R. V. [2011] AI-complete CAPTCHAs as zero knowledge proofs of access to an artificially intelligent system. ISRN Artificial Intelligence, 2012. Google Scholar
Yampolskiy, R. [2012] Leakproofing the singularity artificial intelligence confinement problem, J. Conscious. Stud. 19(1–2), 1–2. Google Scholar
Yampolskiy, R. V. [2013a], Artificial intelligence safety engineering: Why machine ethics is a wrong approach, in Philosophy and Theory of Artificial Intelligence (Springer, Berlin, Heidelberg) pp. 389–396. Crossref, Google Scholar
Yampolskiy, R. V. [2013b] Efficiency Theory: A unifying theory for information, computation and intelligence, J. Discrete Math. Sci. Cryptogr. 16(4–5), 259–277. Crossref, Google Scholar
Yampolskiy, R. V. [2014] Utility function security in artificially intelligent agents, J. Exp. Theor. Artif. Intel. 26(3), 373–389. Crossref, Google Scholar
Yampolskiy, R. V. [2015] Artificial Superintelligence: A Futuristic Approach (Chapman and Hall/CRC). Crossref, Google Scholar
Yampolskiy, R. V. [2016a] On the origin of synthetic life: attribution of output to a particular algorithm, Phys. Scrip. 92(1), 013002. Crossref, Google Scholar
Yampolskiy, R. V. [2016b] Taxonomy of pathways to dangerous artificial intelligence, in Paper presented at the AAAI Workshop: AI, Ethics, and Society. Google Scholar
Yampolskiy, R. V. [2017a] Detecting Qualia in Natural and Artificial Agents. arXiv preprint arXiv:1712.04020. Google Scholar
Yampolskiy, R. [2017b] Job ad: universe designers. In C. Chase (Ed.), Stories from 2045 (pp. 50–53). Google Scholar
Yampolskiy, R. V. [2018] Artificial Intelligence Safety and Security (CRC Press). Crossref, Google Scholar
Yudkowsky, E. [2004] Coherent extrapolated volition. Singularity Institute for Artificial Intelligence. Google Scholar
Zhavoronkov, A. [December 12, 2018] Is Life A Recursive Video Game? Paper presented at the Forbes, Available at: https://www.forbes.com/sites/cognitiveworld/2018/12/12/is-life-a-recursive-video-game. Google Scholar
Ziesche, S. and Yampolskiy, R. V. [2016]. Artificial Fun: Mapping Minds to the Space of Fun. arXiv preprint arXiv:1606.07092. Google Scholar