Hledání novosti a diverzita kvality v otevřené evoluci

• Název práce v češtině: Hledání novosti a diverzita kvality v otevřené evoluci
• Název v anglickém jazyce: Novelty search and quality diversity in open-ended evolution
• Klíčová slova: hledání novosti|diverzita kvality|otevřená evoluce|evoluční algoritmy
• Klíčová slova anglicky: novelty search|quality diversity|open-ended evolution|evolutionary algorithms
• Akademický rok vypsání: 2022/2023
• Typ práce: diplomová práce
• Ústav: Katedra teoretické informatiky a matematické logiky (32-KTIML)
• Vedoucí / školitel: Mgr. Roman Neruda, CSc.
• Řešitel: skrytý - zadáno a potvrzeno stud. odd.
• Datum přihlášení: 16.09.2023
• Datum zadání: 16.09.2023
• Datum potvrzení stud. oddělením: 07.12.2023

Zásady pro vypracování

Open-ended evolution represents a research field combining ideas of evolutionary algorithms and artificial life. The focus is on designing a system that can continuously evolve and change, rather than reaching a stable state. The novelty search and quality diversity methods from evolutionary algorithms are natural candidates for designing open-ended evolving systems.

The goal of the thesis is to study open-endedness in the task of evolution of structures in the Minecraft environment provided by the EvoCraft API. The student will propose and design an open-ended algorithm utilizing appropriate encoding of individuals that supports complexity and adaptation. Mechanisms of novelty search and quality diversity should be utilized in the design of the algorithm. Other desired features of the approach include constraints on time and resources, and interaction of the individuals with the environment. The success of the algorithm will be measured by standard criteria of population diversity, evolution dynamics, divergence, and increasing complexity of the solutions. The proposed algorithms will be realized in the Python programming language and used for experiments demonstrating the soundness of the proposal.

Seznam odborné literatury

Grbic, D., Palm, R.B., Najarro, E., Glanois, C., Risi, S. (2021). EvoCraft: A New Challenge for Open-Endedness. In: Castillo, P.A., Jiménez Laredo, J.L. (eds) Applications of Evolutionary Computation. EvoApplications 2021. Lecture Notes in Computer Science, vol 12694. Springer, Cham. https://doi.org/10.1007/978-3-030-72699-7_21

Justin K. Pugh, L. B. Soros, Rafaela Frota, Kevin Negy, Kenneth O. Stanley; September 4–8, 2017. "Major evolutionary transitions in the Voxelbuild virtual sandbox game." Proceedings of the ECAL 2017, the Fourteenth European Conference on Artificial Life. ECAL 2017, the Fourteenth European Conference on Artificial Life. Lyon, France. (pp. 553-560). ASME. https://doi.org/10.1162/isal_a_088

Suarez, Joseph & Du, Yilun & Isola, Phillip & Mordatch, Igor. (2019). Neural MMO: A Massively Multiagent Game Environment for Training and Evaluating Intelligent Agents. arXiv:1903.00784v1 [cs.MA]. https://arxiv.org/pdf/1903.00784.pdf

Alyssa M Adams, Sara I Walker. (2018). Real-world open-ended evolution: a League of Legends adventure. International Journal of Design & Nature and Ecodynamics.
Vol. 12, Issue 4, 458-469. WIT Press. https://www.witpress.com/elibrary/dne-volumes/12/4/1768

Corominas-Murtra Bernat, Seoane Luís F. and Solé Ricard . (2018). Zipf’s Law, unbounded complexity and open-ended evolution. Journal of. Royal Soc. Interface. 15: 20180395. http://doi.org/10.1098/rsif.2018.0395

Pio-Lopez L, Bischof J, LaPalme JV, Levin M. (2023) The scaling of goals from cellular to anatomical homeostasis: an evolutionary simulation, experiment and analysis. Journal of. Royal Soc. Interface 13: 20220072. https://doi.org/10.1098/rsfs.2022.0072

Lehman, J. and Stanley, K. (2011). Abandoning objectives: Evolution through the search for novelty alone. Evolutionary computation, 19:189–223.

Meyerson, E., Lehman, J., and Miikkulainen, R. (2016). Learning behavior characterizations for novelty search. In Proceedings of the Genetic and Evolutionary Computation Conference 2016, GECCO ’16, page 149–156, New York, NY, USA. Association for Computing Machinery.