… of active Brownian particles, activeparticles undergoing … in coupling machine-learningmodels to activeparticles, in a quest to … -learning approach, ActiveNet, that can be trained to learn …
… With the help of a tabular Q-learning algorithm, we design a model for training a … We show that the strategy obtained with our reinforcement learningmodel indeed represents an efficient …
A Zöttl, H Stark - Annual Review of Condensed Matter Physics, 2023 - annualreviews.org
… colloidal engines, adaptive responses of active colloids with the help of machine learning … on modelingactive matter on the particle level and not on field-based continuum modeling, …
This paper proposes a systems approach to the theory of perception and learning in populations composed of many living entities. Starting from a phenomenological description of …
M Nasiri, E Loran, B Liebchen - Proceedings of the …, 2024 - National Acad Sciences
… To date, it remains an open problem to understand how well-established foraging models such as Lévy walks or chemotaxis models would hold up when we account for key real-world …
… model-based reinforcement learning [150,151] and more importantly world models [152] where the agents strive to learn a model (… can even be translated to learning the physics of the …
… issue: Can we identify the active species in a binary model system of active and passive … -based machine learning approach for the identification of activeparticles in an active/passive …
… tools witness the interest of applied mathematicians toward modeling, qualitative analysis, and computing of large systems of activeparticles viewed as living, hence complex, systems. …
M Nasiri, B Liebchen - New Journal of Physics, 2022 - iopscience.iop.org
… Here we develop a machine learning-based approach that … -based deep reinforcement learning techniques and, crucially, … a universal path planner for future intelligent activeparticles. …