Learning of Population Dynamics: Inverse Optimization Meets JKO Scheme
By: Mikhail Persiianov , Jiawei Chen , Petr Mokrov and more
Potential Business Impact:
Helps scientists track how groups of things change.
Learning population dynamics involves recovering the underlying process that governs particle evolution, given evolutionary snapshots of samples at discrete time points. Recent methods frame this as an energy minimization problem in probability space and leverage the celebrated JKO scheme for efficient time discretization. In this work, we introduce $\texttt{iJKOnet}$, an approach that combines the JKO framework with inverse optimization techniques to learn population dynamics. Our method relies on a conventional $\textit{end-to-end}$ adversarial training procedure and does not require restrictive architectural choices, e.g., input-convex neural networks. We establish theoretical guarantees for our methodology and demonstrate improved performance over prior JKO-based methods.
Similar Papers
Convergence of the fully discrete JKO scheme
Analysis of PDEs
Makes computer models of moving crowds more accurate.
Learning from Samples: Inverse Problems over measures via Sharpened Fenchel-Young Losses
Machine Learning (CS)
Helps understand how things change over time.
An Evolutionary Multi-objective Optimization for Replica-Exchange-based Physics-informed Operator Learning Network
Machine Learning (CS)
Solves hard math problems with less data.