“…Great progress has been made in the applications of deep learning technology for seismic data processing and interpretation, such as travel time computation (Waheed et al., 2021), seismic trace interpolation (Wang et al., 2020), first‐break picking (Yuan et al., 2018), velocity model building (Chen & Schuster, 2020; Chen & Saygin, 2021; Yang & Ma, 2019; Yu & Ma, 2021), geological body recognition (Huang et al., 2017), seismic facies classification (Ross & Cole, 2017), reservoir characterization (Abid et al., 2021), passive seismic event detection (Othman et al., 2022) and microseismic monitoring (Shaheen et al., 2021). For seismic data denoising, the convolutional neural network (CNN), a commonly used deep learning network (LeCun et al., 2015), has made good performance in random noise attenuation (Liu et al., 2020; Wang & Chen, 2019; Xie et al., 2018; Yu et al., 2019; Zhao et al., 2019b), such as white noise (Wu et al., 2019) and swell noise (Zhao et al., 2019a; You et al., 2020).…”