Discover the Mysteries of the Maya: Selected Contributions from the Machine Learning Challenge & The Discovery Challenge Workshop at ECML PKDD 2021

  title={Discover the Mysteries of the Maya: Selected Contributions from the Machine Learning Challenge \& The Discovery Challenge Workshop at ECML PKDD 2021},
  author={Dragi Kocev and Nikola Simidjievski and Ana Kostovska and Ivica Dimitrovski and Žiga Kokalj},
Exploration of the Maya forest region remotely through machine learning has recently accelerated. Using experts to manually look at satellite data is time-consuming and expensive. The machine learning competition Discover the mysteries of the Maya addresses this problem and calls for a competition to improve the performance of state-of-the-art models to automatically detect objects of interest using satellite images. With a given LiDAR image, the model should detect three classes of objects… 



ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data

Billion-scale semi-supervised learning for image classification

This paper proposes a pipeline, based on a teacher/student paradigm, that leverages a large collection of unlabelled images to improve the performance for a given target architecture, like ResNet-50 or ResNext.

Learning to Classify Structures in ALS-Derived Visualizations of Ancient Maya Settlements with CNN

This work considered several variations of the VGG-19 Convolutional Neural Network to solve the task of classifying visualized example structures from previously manually annotated ALS images of man-made aguadas, buildings and platforms, as well as images of surrounding terrain (four classes and over 12,000 anthropogenic structures).

Semantic Segmentation of Airborne LiDAR Data in Maya Archaeology

This work has built and compared two deep learning-based models, U-Net and Mask R-CNN, for semantic segmentation, and outlines the value of these models for archaeological research and presents the road map to produce a useful decision support system for recognition of ancient objects in LiDAR data.

Self-Attention Generative Adversarial Networks

The proposed SAGAN achieves the state-of-the-art results, boosting the best published Inception score from 36.8 to 52.52 and reducing Frechet Inception distance from 27.62 to 18.65 on the challenging ImageNet dataset.

A disciplined approach to neural network hyper-parameters: Part 1 - learning rate, batch size, momentum, and weight decay

This report shows how to examine the training validation/test loss function for subtle clues of underfitting and overfitting and suggests guidelines for moving toward the optimal balance point and discusses how to increase/decrease the learning rate/momentum to speed up training.

Learning to Look at LiDAR: The Use of R-CNN in the Automated Detection of Archaeological Objects in LiDAR Data from the Netherlands

A promising new technique for the automated detection of multiple classes of archaeological objects in LiDAR data is presented, based on R-CNNs (Regions-based Convolutional Neural Networks), which is able to automatically detect and categorise these two types of archaeologists objects.

Deep Residual Learning for Image Recognition

This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.

U-Net: Convolutional Networks for Biomedical Image Segmentation

It is shown that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks.