Alberto Tonda
Isabelle Alvarez
Giovanni Squillero
ABSTRACT
The mathematical theory of viability, developed to formalize problems related to natural and social phenomena, investigates the evolution of dynamical systems under constraints. A main objective of this theory is to design control laws to keep systems inside viable domains. Control laws are traditionally defined as rules, based on the current position in the state space with respect to the boundaries of the viability kernel. However, finding these boundaries is a computationally expensive procedure, feasible only for trivial systems. We propose an approach based on Genetic Programming (GP) to discover control laws for viability problems in analytic form. Such laws could keep a system viable without the need of computing its viability kernel, facilitate communication with stakeholders, and improve explainability. A candidate set of control rules is encoded as GP trees describing equations. Evaluation is noisy, due to stochastic sampling: initial conditions are randomly drawn from the state space of the problem, and for each, a system of differential equations describing the system is solved, creating a trajectory. Candidate control laws are rewarded for keeping viable as many trajectories as possible, for as long as possible. The proposed approach is evaluated on established benchmarks for viability and delivers promising results.
- Isabelle Alvarez and Sophie Martin. 2011. Geometric robustness of viability kernels and resilience basins. In Viability and Resilience of Complex Systems: Concepts, Methods and Case Studies from Ecology and Society. Guillaume Deffuant and Nigel Gilbert, (Eds.) Springer Berlin Heidelberg, Berlin, Heidelberg, 193--218.Google Scholar
- Isabelle Alvarez, Laetitia Zaleski, Jean-Pierre Briot, and Marta de A. Irving. 2023. Collective management of environmental commons with multiple usages: a guaranteed viability approach. Ecological Modelling, 475, 110186. Google ScholarCross Ref
- J.-P. Aubin. 1991. Viability Theory. Birkhäuser, Basel.Google Scholar
- Jean Pierre Aubin. 1997. Dynamic economic theory: a viability approach. Vol. 5. Springer Verlag.Google Scholar
- Jean-Pierre Aubin, Alexandre Bayen, and Patrick Saint-Pierre. 2011. Viability Theory: New Directions. Springer, 830. Google ScholarCross Ref
- Jean-Pierre Aubin and Halina Frankowska. 1985. Heavy viable trajectories of controlled systems. Annales de l'Institut Henri Poincaré C, Analyse non linéaire, 2, 5, 371--395. Google ScholarCross Ref
- C. Bernard and S. Martin. 2012. Building strategies to ensure language coexistence in presence of bilingualism. Applied Mathematics and Computation, 218, 17, 8825--8841. Google ScholarCross Ref
- Erick Cantú-Paz. 2004. Adaptive sampling for noisy problems. Tech. rep. Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States).Google Scholar
- S. Carpenter, W. Brock, and P. Hanson. 1999. Ecological and social dynamics in simple models of ecosystem management. Conservation Ecology, 3(2). http://www.consecol.org/vol3/iss2/art4/.Google Scholar
- Duc-Cuong Dang and Per Kristian Lehre. 2015. Efficient Optimisation of Noisy Fitness Functions with Population-based Evolutionary Algorithms. en. In Proceedings of the 2015 ACM Conference on Foundations of Genetic Algorithms XIII. ACM, Aberystwyth United Kingdom, (Jan. 2015), 62--68. isbn: 978-1-4503-3434-1. Google ScholarDigital Library
- M. Falcone and P. Saint-Pierre. 1987. Slow and quasi-slow solutions of differential inclusions. Nonlinear Analysis: Theory, Methods & Applications, 11, 3, 367--377. Google ScholarDigital Library
- J. Michael Fitzpatrick and John J. Grefenstette. 1988. Genetic Algorithms in Noisy Environments. Machine Learning, 3, 2/3, 101--120. 4003254. Google ScholarCross Ref
- Marco Gaudesi, Elio Piccolo, Giovanni Squillero, and Alberto Tonda. 2016. Exploiting evolutionary modeling to prevail in iterated prisoner's dilemma tournaments. IEEE Transactions on Computational Intelligence and AI in Games, 8, 3, (Sept. 2016), 288--300. Google ScholarCross Ref
- David Goldberg E., Kalyanmoy Deb, and James H. Clark. 1992. Genetic algorithms, noise, and the sizing of populations. Complex Systems, 6, 333--362.Google Scholar
- John J. Grefenstette and J. Michael Fitzpatrick. 1985. Genetic search with approximate function evaluation. In Proceedings of the 1st International Conference on Genetic Algorithms. L. Erlbaum Associates Inc., USA, 112--120. isbn: 0805804269.Google Scholar
- E. Hairer, S. P. Norsett, and G. Wanner. 1993. Solving Ordinary Differential Equations I: Nonstiff problems. Springer Berlin Heidelberg. Google ScholarCross Ref
- Baligh Al-Helali, Qi Chen, Bing Xue, and Mengjie Zhang. 2020. Genetic Programming with Noise Sensitivity for Imputation Predictor Selection in Symbolic Regression with Incomplete Data. In 2020 IEEE Congress on Evolutionary Computation (CEC). IEEE, Glasgow, United Kingdom, (July 2020), 1--8. isbn: 978-1-72816-929-3. Google ScholarDigital Library
- Torsten Hildebrandt and Jürgen Branke. 2015. On Using Surrogates with Genetic Programming. Evolutionary Computation, 23, 3, (Sept. 2015), 343--367. Google ScholarDigital Library
- Gregory Hornby, Al Globus, Derek Linden, and Jason Lohn. 2006. Automated antenna design with evolutionary algorithms. In Space 2006. American Institute of Aeronautics and Astronautics, (Sept. 2006). Google ScholarCross Ref
- Yaochu Jin and J. Branke. 2005. Evolutionary optimization in uncertain environments - a survey. IEEE Transactions on Evolutionary Computation, 9, 3, 303--317. Google ScholarDigital Library
- John R Koza. 1994. Genetic programming as a means for programming computers by natural selection. Statistics and computing, 4, 87--112.Google Scholar
- Fiacc Larkin and Conor Ryan. 2009. Avoiding the pitfalls of noisy fitness functions with genetic algorithms. en. In Proceedings of the 11th Annual conference on Genetic and evolutionary computation. ACM, Montreal Québec Canada, (July 2009), 1861--1862. isbn: 978-1-60558-325-9. Google ScholarDigital Library
- Benoit Leblanc and Evelyne Lutton. 1998. Bitwise regularity and ga-hardness. In ICEC 98, May 5--9, Anchorage, Alaska.Google Scholar
- E. Lutton and J. Lévy Véhel. 1998. Hölder functions and deception of genetic algorithms. IEEE transactions on Evolutionary computation, 2, 2, (July 1998), 56--72.Google ScholarDigital Library
- E. Lutton and J. Lévy Véhel. 2006. Pointwise regularity of fitness landscapes and the performance of a simple es. In CEC'06. Vancouver, Canada, (July 2006).Google Scholar
- Jordan MacLachlan, Yi Mei, Juergen Branke, and Mengjie Zhang. 2020. Genetic Programming Hyper-Heuristics with Vehicle Collaboration for Uncertain Capacitated Arc Routing Problems. Evolutionary Computation, 28, 4, (Dec. 2020), 563--593. Google ScholarDigital Library
- S. Martin. 2004. The cost of restoration as a way of defining resilience: a viability approach applied to a model of lake eutrophication. Ecology and Society, 9, 2, 19. https://hal.inrae.fr/hal-02583472.Google ScholarCross Ref
- S. Martin and I. Alvarez. 2019. Anticipating shocks in the state space: characterizing robustness and building increasingly robust evolutions. SIAM Journal on Control and Optimization, 57, 1, 490--509. doi: 10.1137/16M1061175. Google ScholarDigital Library
- JD. Mathias, J. Anderies, and Janssen. 2017. On our rapidly shrinking capacity to comply with the planetary boundaries on climate change. Scientific Report, 7, 42061. Google ScholarCross Ref
- Salma Mesmoudi, Nathalie Perrot, Romain Reuillon, Paul Bourgine, and Evelyne Lutton. 2010. Optimal viable path search for a cheese ripening process using a multi-objective ea. In ICEC 2010, International Conference on Evolutionary Computation. 24--26 oct, Valencia, Spain. (Oct. 2010).Google Scholar
- Aaron Meurer et al. 2017. Sympy: symbolic computing in python. PeerJ Computer Science, 3, (Jan. 2017), e103. Google ScholarCross Ref
- Brad L. Miller and David E. Goldberg. 1995. Genetic algorithms, tournament selection, and the effects of noise. Complex Syst., 9.Google Scholar
- Luis F. Miranda, Luiz Otavio V. B. Oliveira, Joao Francisco B. S. Martins, and Gisele L. Pappa. 2017. How noisy data affects geometric semantic genetic programming. In Proceedings of the Genetic and Evolutionary Computation Conference (GECCO '17). Association for Computing Machinery, Berlin, Germany, 985--992. isbn: 9781450349208. Google ScholarDigital Library
- Aïchouche Oubraham and Georges Zaccour. 2018. A survey of applications of viability theory to the sustainable exploitation of renewable resources. Ecological Economics, 145, 346--367. Google ScholarCross Ref
- Alain Ratle. 1998. Accelerating the convergence of evolutionary algorithms by fitness landscape approximation. In Proceedings of the 5th International Conference on Parallel Problem Solving from Nature (PPSN V). Springer-Verlag, Berlin, Heidelberg, 87--96. isbn: 3540650784.Google ScholarCross Ref
- Magnus Rattray and Jonathan Shapiro. 1996. Noisy fitness evaluation in genetic algorithms and the dynamics of learning. In Proceedings of the 4th Workshop on Foundations of Genetic Algorithms. San Diego, CA, USA, August 5 1996. Richard K. Belew and Michael D. Vose, (Eds.) Morgan Kaufmann, 117--139.Google Scholar
- Patrick Saint-Pierre. 1994. Approximation of the viability kernel. Applied Mathematics and Optimization, 29, 2, 187--209.Google ScholarDigital Library
- Yasuhito Sano and Hajime Kita. 2002. Optimization of Noisy Fitness Functions by means of Genetic Algorithms using History of Search. en. IEEJ Transactions on Electronics, Information and Systems, 122, 6, 1001--1008. 1987.122.6_1001. Google ScholarCross Ref
- Michael Schmidt and Hod Lipson. 2009. Distilling free-form natural laws from experimental data. science, 324, 5923, 81--85.Google Scholar
- Sara Silva, Leonardo Vanneschi, Ana I.R. Cabral, and Maria J. Vasconcelos. 2018. A semi-supervised genetic programming method for dealing with noisy labels and hidden overfitting. Swarm and Evolutionary Computation, 39, 323--338. Google ScholarCross Ref
- Fatima-Zahra Tani, Alain Rapaport, and Térence Bayen. 2019. A hybrid control against species invasion in the chemostat. In 2019 IEEE 58th Conference on Decision and Control (CDC), 2814--2819. Google ScholarDigital Library
- Pauli Virtanen et al. 2020. SciPy 1.0: Fundamental Algorithms for Scientific Computing in Python. Nature Methods, 17, 261--272. 686-2. Google ScholarCross Ref
Index Terms
- Towards Evolutionary Control Laws for Viability Problems
Recommendations
Micro/macro viability analysis of individual-based models: Investigation into the viability of a stylized agricultural cooperative
The mathematical viability theory proposes methods and tools to study at a global level how controlled dynamical systems can be confined in a desirable subset of the state space. Multilevel viability problems are rarely studied since they induce ...
A Comparison of three evolutionary strategies for multiobjective genetic programming
We report what we believe to be the first comparative study of multi-objective genetic programming (GP) algorithms on benchmark symbolic regression and machine learning problems. We compare the Strength Pareto Evolutionary Algorithm (SPEA2), the Non-...
Population viability in three trophic-level food chains
The perpetuation of three-trophic level ecosystems where the three species exhibit unpredictable time-varying survival strategies is described by a specific set, the viability kernel, gathering all states from which there exists at least one trajectory ...
Comments