A new method for parameter estimation in probabilistic models: Minimum probability flow
This addresses the challenge of parameter estimation in probabilistic models for researchers and practitioners, offering a more efficient and accurate method, though it appears incremental as it builds on existing fitting techniques.
The authors tackled the problem of fitting probabilistic models to data, which is often difficult due to intractable partition functions, by proposing Minimum Probability Flow (MPF) as a new parameter fitting method applicable to any parametric model. They demonstrated that in an Ising spin glass case, MPF outperformed current techniques by at least an order of magnitude in convergence time with lower error in recovered coupling parameters.
Fitting probabilistic models to data is often difficult, due to the general intractability of the partition function. We propose a new parameter fitting method, Minimum Probability Flow (MPF), which is applicable to any parametric model. We demonstrate parameter estimation using MPF in two cases: a continuous state space model, and an Ising spin glass. In the latter case it outperforms current techniques by at least an order of magnitude in convergence time with lower error in the recovered coupling parameters.