Deep active learning aims to reduce the annotation cost for the training of deep models, which is notoriously data-hungry. Until recently, deep active learning methods were ineffectual inthelow-budgetregime, where only asmall number ofexamples areannotated. Thesituation hasbeen alleviated byrecent advances inrepresentation andself-supervised learning, which impart thegeometry ofthe data representation with rich information about the points.

Since many areas, such as biology or discrete-event systems, are naturally described in continuous time, we present a model based on a Markov jumpprocessmodulating asubordinated diffusionprocess. Weprovidetheexact evolution equations fortheprior andposterior marginal densities, thedirect solutions of which are however computationally intractable.

Spiking neural networks (SNNs) represent a promising approach to developing artificial neural networks that are both energy-efficient and biologically plausible.