Machine learning-assisted advances in graphene and 2D materials

Machine learning (ML) is becoming an enabling layer for graphene and two-dimensional (2D) material research, helping to manage the growing complexity of spectroscopy, electron transport, and multiscale simulation data. This brief review surveys recent advances across four representative domains: (i) ML-assisted characterization of defects and nanostructures, (ii) ML-based inference of local material properties, (iii) ML approaches for interpreting phase-coherent quantum transport in mesoscopic devices, and (iv) ML-accelerated simulations/modeling and inverse-design workflows. Across these areas, ML is highlighted as a complementary methodology that enhances parameter inference and automates analysis in regimes where conventional approaches are limited. We conclude by outlining challenges for reliable deployment of ML methods in graphene and 2D materials research, and by discussing promising directions toward more robust ML-integrated discovery and design in graphene and related 2D material platforms. • Reviews ML-assisted advances in graphene and 2D materials research. • Covers defect analysis, local inference, quantum transport, and modeling. • Emphasizes physics-guided interpretation beyond black-box prediction. • Highlights ML for spectroscopy, microscopy, and multiscale simulations. • Outlines challenges in data quality, transferability, and interpretability.