Worldwide, several collaborations are working on the extension of our knowledge about neutrino masses, mixing angles and the CP violating phase. The analyses of neutrino-oscillation experiments however suffer from the tension between the need for fast Monte-Carlo simulations and for a microscopic description of the nuclear processes governing the neutrino-nucleus interactions. The goal of this project is to combine the expertise of the Ghent group in microscopic modeling with the experience of the Geneva partner in experimental analyses using generators. We will join forces to develop methods for efficient multivariate MC generation based on neural networks, of direct use in neutrino-oscillation analyses. The recent developments we have made in Artificial Intelligence techniques suitable for particle physics applications will be pivotal in this effort. In recent work we have shown that modern Neural Net technology are able to reduce the time needed to sample complex probability density functions considerably. The combination of detailed neutrino-nucleus interaction models (Ghent) with advanced numerical models (Geneva) will propel the understanding of these interactions forward and will strongly impact the sensitivity of the search for particle-antiparticle asymmetries and other beyond-standard-model phenomena in neutrino oscillations. In this way we will be able to make strong contributions to the future successes in this field.