Machine learning left-right breaking from gravitational waves

Abstract First-order phase transitions in the early universe can generate stochastic gravitational waves (GWs), providing a unique probe of high-scale particle physics. The Left-Right Symmetric Model (LRSM), which restores parity symmetry at high energies and naturally incorporates the seesaw mechanism, allows for such transitions — particularly during the spontaneous breaking of SU(2) R × SU(2) L × U(1) B-L → SU(2) L × U(1) Y . This initial step, though less studied, is theoretically motivated and can produce observable GW signals. In this work, we investigate the GW signatures of this first-step phase transition in the minimal LRSM using the high-precision three-dimensional effective field theory framework via PhaseTracer and DRalgo . We identify regions of parameter space that give rise to strongly first-order transitions with GW spectra detectable at forthcoming observatories such as BBO and DECIGO. To efficiently explore the high-dimensional parameter space, we also employ a Machine Learning Scan strategy and perform a sensitivity analysis to determine which parameters most strongly influence the GW predictions.