Abstract: Disclosed is a method performed by an access point, AP (120), of a WLAN (125), for handling a wireless device (140) connected to the AP (120). The method comprises obtaining information on signal quality of signals sent between the AP (120) and the wireless device (140), and when the obtained information on signal quality reveals that the signal quality is below a signal quality threshold, determining a current value of a wireless device characteristic. The method further comprises obtaining, from a pre-emptive handover system (150), an estimation of a likelihood of the wireless device (140) disconnecting from the WLAN for the current value of the wireless device characteristic, wherein the likelihood was determined from statistical information of earlier handling of the wireless device (140) by the AP (120) for different values of the wireless device characteristics.

Abstract: A method for enabling a machine learning model to be aggregated from local machine learning models comprised in at least two local nodes in a wireless communications network is provided. The method comprises receiving, from each of the at least two local nodes, a parametrized function of a local machine learning model, a generator function of a local generative model, and a discriminator function of a local discriminative model, wherein the generator function and the discriminator function are trained on the same data as the parametrized function.