Invasive species are significant contributors to global changes and constitute a severe threat to biodiversity. Yet invasions offer an incredible framework to understand how small and low-diverse introduced populations adapt to novel environmental conditions and succeed in colonizing large areas. However, due to the insufficient data on the origin of the first introduced propagule and the first stage of invasion, reconstructing a species’ invasion history is challenging. Here, we applied genetic clustering methods and explicit admixture tests combined with ABC models and Machine Learning algorithms to describe the phylogeography of native and invasive populations and infer the most probable demographic invasion scenarios of Pseudorasbora parva, a highly invasive freshwater fish and the healthy carrier of a novel lethal fungi-like pathogen (Sphaerothecum destruens), which is responsible for the decline of several fish species in Europe. We found that the current genetic structuring of the native P. parva range has been shaped by waves of gene flow originating from southern and northern Chinese populations. Furthermore, our results strongly suggest that the invasive genetic diversity is the outcome of past recurrent global invasion pathways of admixed native populations. Our study also illustrates how the combination of admixture tests, ABC, Machine Learning can be used to detect high-resolution demographic signatures and reconstruct an integrative biological invasion history.