Research Interests

My research is centered on brain-inspired machine learning, with a focus on latent dynamical modeling, spatiotemporal representation learning, and manifold learning in latent space. I am motivated by how intelligent systems can learn structured internal representations of complex environments—capturing temporal evolution, spatial dependencies, and geometric structure. My current work targets structured latent dynamics for neural signals (BOLD signals and functional connectivity), exploring constrained Koopman operators and geometric structure for interpretable, stable representations. I aim to leverage machine learning models to learn latent-space representations that capture both neural dynamics and underlying manifold structure, developing frameworks that bridge neural data analysis with questions of representation, adaptation, and brain-inspired intelligence.