Path-filtering in path-integral simulations of open quantum systems using GFlowNets.
Autor: | Lackman-Mincoff J; Department of Chemistry, McGill University, Montreal, Quebec H3A 0G4, Canada., Jain M; Mila, University of Montreal, Montreal, Quebec H2S 3H1, Canada., Malkin N; Mila, University of Montreal, Montreal, Quebec H2S 3H1, Canada., Bengio Y; Mila, University of Montreal, Montreal, Quebec H2S 3H1, Canada., Simine L; Department of Chemistry, McGill University, Montreal, Quebec H3A 0G4, Canada. |
---|---|
Jazyk: | angličtina |
Zdroj: | The Journal of chemical physics [J Chem Phys] 2024 Oct 14; Vol. 161 (14). |
DOI: | 10.1063/5.0226408 |
Abstrakt: | An important class of methods for modeling dynamics in open quantum systems is based on the well-known influence functional (IF) approach to solving path-integral equations of motion. Within this paradigm, path-filtering schemes based on the removal of IF elements that fall below a certain threshold aim to reduce the effort needed to calculate and store the influence functional, making very challenging simulations possible. A filtering protocol of this type is considered acceptable as long as the simulation remains mathematically stable. This, however, does not guarantee that the approximated dynamics preserve the physics of the simulated process. In this paper, we explore the possibility of training Generative Flow Networks (GFlowNets) to produce filtering protocols while optimizing for mathematical stability and for physical accuracy. Trained using the trajectory balance objective, the model produces sets of paths to be added to a truncated initial set; it is rewarded if the combined set of paths gives rise to solutions in which the trace of the density matrix is conserved, the populations remain real, and the dynamics approach the exact reference. Using a simple two-level system coupled to a dissipative reservoir, we perform proof-of-concept simulations and demonstrate the elegant and surprising filtering solutions proposed by the GFlowNet. (© 2024 Author(s). Published under an exclusive license by AIP Publishing.) |
Databáze: | MEDLINE |
Externí odkaz: |