This work studies the decoherence of chiral magnon edge modes in a topological bosonic Chern insulator. The paper appears in the Focus Issue on Topological Physics: From Fundamentals to Applications in the Journal of Physics: Condensed Matter. We analyze how the topologically protected edge modes lose quantum coherence due to environmental coupling, and characterize the timescales and mechanisms of decoherence relevant to quantum magnonic devices.
We present a data-driven methodology for pipeline leak identification and severity analysis, combining acoustic sensing with deep learning models. The approach uses time-series acoustic data collected along industrial pipelines, processed through convolutional and recurrent neural network architectures for leak localization and severity classification. Validation on real-world pipeline data demonstrates high accuracy in both tasks. This work arose from the TCS Research and Innovation Fellowship at IIT Kharagpur Research Park.