A detailed understanding of the influence of long-term accumulation of plastic deformations on fatigue crack initiation behavior of railway rails requires simulating their entire service life, from virgin to highly deformed material states. This involves finite element simulations of rails subjected to many wheel passages, which can result in significant computational time when using time-domain material models. To speed up the simulation of cyclic loading, Suiker and de Borst [1] proposed a cycle-domain model to simulate the evolution of maximum plastic deformations in ballasted tracks subjected to many loading cycles. The model framework is based on standard plasticity theory and is formulated as a viscoplastic model [1]. In this contribution, inspired by [1], we substitute a standard time-domain plasticity model, specifically the Chaboche model, with a cycle-domain model, where time derivatives of quantities are replaced by their change per unit cycle. The proposed approach is to study the cycle-domain model for isotropic material under uniaxial loading and to calibrate the model using a neural network. Training data is obtained using the Chaboche model with different strain amplitudes. Finally, symbolic regression is adopted to discover an analytical expression for the norm of plastic strain increment, and the results are compared against those from the neural network model. REFERENCES [1] Suiker A.S., and de Borst R., A numerical model for the cyclic deterioration of railway tracks. International journal for numerical methods in engineering, Vol. 57 (4), pp. 441-470, 2003.