Reinforcement Learning

The Reinforcement Learning project, led by Juergen Schmidhuber at IDSIA aimed to investigate agent interactions in dynamic environments. The project further extended and applied state-of-the-art results on practical and theoretical aspects of general reinforcement learners and search/optimization algorithms. My contribution to this project was the study of a Fuzzy Lattice Reasoning (FLR) classifier for rule induction (learning) and fuzzy reasoning (generalization).

An open-source implementation has been contributed to the Waikato Environment for Knowledge Analysis (WEKA).

Related publications:

• O.G. Younis, L. Corinzia, I. N. Athanasiadis, A. Krause, J. M. Buhmann, M. Turchetta, Breeding Programs Optimization with Reinforcement Learning, NeurIPS 2023 Workshop on Tackling Climate Change with Machine Learning, 2023, doi:10.48550/arXiv.2406.03932.
• M.G. Kallenberg, H. Overweg, R. van Bree, I.N. Athanasiadis, Nitrogen management with reinforcement learning and crop growth models, Environmental Data Science, 2:e34, 2023, Cambridge University Press, doi:10.1017/eds.2023.28.
• M. Turchetta, L. Corinzia, S. Sussex, A. Burton, J. Herrera, I. N. Athanasiadis, J. M. Buhmann, A. Krause, Learning long-term crop management strategies with CyclesGym, Thirty-sixth Conference on Neural Information Processing Systems Datasets and Benchmarks Track, 2022.
• H. Overweg, H. N. C. Berghuijs, I. N. Athanasiadis, CropGym: a Reinforcement Learning Environment for Crop Management, Computing Research Repository, AIMOCC 2021 Workshop at 9th Int'l Conf Learning Representations (ICLR 2021), 2021, doi:10.48550/arXiv.2104.04326.
• V. G. Kaburlasos, I. N. Athanasiadis, P. A. Mitkas, Fuzzy Lattice Reasoning (FLR) classifier and its application for ambient ozone estimation, International Journal of Approximate Reasoning, 45:152-188, 2007, doi:10.1016/j.ijar.2006.08.001.