“…This strategy includes methods based on the Fourier transform (Mousavi and Langston, 2016a), wavelet transform (Mousavi and Langston, 2016b; Mousavi et al ., 2016), seislet transform (Fomel and Liu, 2010; Chen et al ., 2014), curvelet transform (Naghizadeh and Sacchi, 2010), shearlet transform (Liu et al ., 2016), Radon transform (Beylkin, 1987), wavelet transform (Sweldens, 1995), dictionary learning (Siahsar et al ., 2017a, 2017b; Chen, 2020) and deep‐learning techniques (Li et al ., 2018b; Zhu et al ., 2019). Another type of approach for random‐noise attenuation focuses on enhancing useful signals and suppressing random noise by utilizing a predictable property, such as predictive filtering (Abma and Claerbout, 1995), predictive filtering (Canales, 1984; Gulunay, 1986), the complex‐trace analysis method (Karsli et al ., 2006), non‐stationary predictive filtering (Liu et al ., 2012; Liu and Chen, 2013; Liu et al ., 2018; Li et al ., 2018a) and the polynomial fitting–based approach (Lu and Lu, 2009).…”