 This study aims to develop accurate landslide maps using machine learning methods such as Artificial Neural Network ANN, Support Vector Machines, SVM, Random Forest, RF, and Deep Learning Convolution Neural Networks, CNNs. The authors used optical data from the RapidEye satellite and topographic factors to analyze the potential of these methods for landslide detection in the highly landslide-prone Razowa district in Nepal. They created 20 different maps using ANN, SVM, RF, and different CNN instantiations and compared their results with extensive fieldwork through mean intersection over union MIU, and other common metrics. The study found that the small window-sized CNN using spectral information only achieved the best result of 78.26% MIU. However, the authors concluded that CNNs do not automatically outperform ANN, SVM, and RF, and their performance strongly depends on their design, input window sizes, and training strategies. The study suggests that deep learning can improve landslide mapping in the future if the effects of different designs are better understood, enough training samples exist, and the effects of augmentation strategies to artificially increase the number of existing samples are better understood. This article was authored by Omid Gorbanzadeh, Thomas Blashk, Khalil Golania, and others. We are article.tv, links in the description below.