MESSY Estimation: Maximum-Entropy based Stochastic and Symbolic densitY Estimation

We introduce MESSY estimation, a Maximum-Entropy based Stochastic and Symbolic densitY estimation method. The proposed approach recovers probability density functions symbolically from samples using moments of a Gradient flow in which the ansatz serves as the driving force. In particular, we construct a gradient-based drift-diffusion process that connects samples of the unknown distribution function to a guess symbolic expression. We then show that when the guess distribution has the maximum entropy form, the parameters of this distribution can be found …

construct cs.ai cs.it cs.lg diffusion drift driving entropy flow functions gradient math.it math.st moments probability process samples stat.ml stat.th stochastic