Our paper entitled Efficient recurrent neural network methods for anomalously diffusing single particle short and noisy trajectories, jointly with O. Garibo-i-Orts, A. Baeza-Bosca, and M.A. García March, has been recently published in J. Phys. A: Math. Theor.
In this work we present a data-driven method able to infer the anomalous exponent and to identify the type of anomalous diffusion process behind single, noisy and short trajectories, with good accuracy. This model was used in our participation in the anomalous diffusion (AnDi) challenge. A combination of convolutional and recurrent neural networks was used to achieve state-of-the-art results when compared to methods participating in the AnDi challenge, ranking top 4 in both classification and diffusion exponent regression.
Our method let us work with short, noisy trajectories either in one-, two- or three dimensions generated by the following models: Fractional Brownian Motion, Continuous Time Random Walk, Annealed Transient Time Motion, Scaled Brownian Motion, and Lévy Walks.
You can access to the work here. You can also write me for a complimentary copy.