In the field of Earth Observation (EO), Continual Learning (CL) algorithms have been proposed to deal with large datasets by decomposing them into several subsets and processing them incrementally. The majority of these algorithms assume that data is (a) coming from a single source, and (b) fully labeled. Real-world EO datasets are instead characterized by a large heterogeneity (e.g., coming from aerial, satellite, or drone scenarios), and for the most part they are unlabeled, meaning they can be fully exploited only through the emerging Self-Supervised Learning (SSL) paradigm. For these reasons, in this paper we present a new algorithm for merging SSL and CL for remote sensing applications, that we call Continual Barlow Twins (CBT). It combines the advantages of one of the simplest self-supervision techniques, i.e., Barlow Twins, with the Elastic Weight Consolidation method to avoid catastrophic forgetting. In addition, we evaluate the proposed Continual Self-Supervised Learning approach on a highly heterogeneous EO dataset, showing the effectiveness of this strategy on a novel combination of three almost non-overlapping domains datasets ( airborne Potsdam, satellite US3D, and drone UAVid), on a crucial downstream task in EO, i.e., semantic segmentation. Encouraging results show the superiority of SSL in this setting, and the effectiveness of creating an incremental effective pretrained feature extractor, based on ResNet50, without the need of relying on the complete availability of all the data, with a valuable saving of time and resources.

Scene understanding of satellite and aerial images is a pivotal task in various remote sensing (RS) practices, such as land cover and urban development monitoring. In recent years, neural networks have become a de-facto standard in many of these applications. However, semantic segmentation still remains a challenging task. With respect to other computer vision (CV) areas, in RS large labeled datasets are not very often available, due to their large cost and to the required manpower. On the other hand, self-supervised learning (SSL) is earning more and more interest in CV, reaching state-of-the-art in several tasks. In spite of this, most SSL models, pretrained on huge datasets like ImageNet, do not perform particularly well on RS data. For this reason, we propose a combination of a SSL algorithm (particularly, Online Bag of Words) and a semantic segmentation algorithm, shaped for aerial images (namely, Multistage Attention ResU-Net), to show new encouraging results (i.e., 81.76% mIoU with ResNet-18 backbone) on the ISPRS Vaihingen dataset.

This paper aims to empirically verify if the board of directors’ (BoD) diversity (i.e., gender, age, and nationality) affects firm performance, which we calculate referring to ROE, ROA, and EBITDA margin. So far, scholars do not converge on a single answer about the effects of observable diversity in the boardrooms on corporate performance. Therefore, this study — referring to a significantly bigger sample — applies machine learning models following a data-driven approach based on a three-year (2017–2019) dataset composed of 59,229 Italian small-medium enterprises (SMEs). The analysis conducted shows that board diversity does not impact firm results, either positively or negatively. The lack of a correlation suggests that there is no reason to not appoint females, young people, and foreigners as directors. The involvement of these “minorities”, which, as shown, does not negatively impact economic-financial results, could on the opposite improve firm reputation as well as enhance the intellectual capital, solving in the meantime a social matter

Large-scale “foundation models” have gained traction as a way to leverage the vast amounts of unlabeled remote sensing data collected every day. However, due to the multiplicity of Earth Observation satellites, these models should learn “sensor agnostic” representations, that generalize across sensor characteristics with minimal fine-tuning. This is complicated by data availability, as low-resolution imagery, such as Sentinel-2 and Landsat-8 data, are available in large amounts, while very high-resolution aerial or satellite data is less common. To better leverage multi-sensor data, we introduce cross-sensor self-supervised training and alignment for remote sensing (X-STARS). We design a self-supervised training loss, the Multi-Sensor Alignment Dense loss (MSAD), to align representations across sensors, even with vastly different resolutions, through a contrastive patch-wise mechanism. Our X-STARS can be applied to train models from scratch, or to adapt large models pretrained on e.g. low-resolution EO data to new high-resolution sensors, in a continual pretraining framework. We collect and release MSC-France, a new multi-sensor dataset, on which we train our X-STARS models, then evaluated on seven downstream classification and segmentation tasks. We demonstrate that X-STARS outperforms the state-of-the-art with less data across various conditions of data availability and resolutions.

Geospatial foundation models (GFMs) have emerged as powerful tools for extracting representations from Earth observation (EO) data, but their evaluation remains inconsistent and narrow. Existing works often evaluate on suboptimal downstream datasets and tasks that are often too easy or too narrow, limiting the usefulness of the evaluations to assess the real-world applicability of GFMs. Additionally, there is a distinct lack of diversity in current evaluation protocols, which fail to account for the multiplicity of image resolutions, sensor types, and temporalities, which further complicates the assessment of GFM performance. In particular, most existing benchmarks are geographically biased toward North America and Europe, questioning the global applicability of GFMs. To overcome these challenges, we introduce PANGAEA, a standardized evaluation protocol that covers a diverse set of datasets, tasks, resolutions, sensor modalities, and temporalities. It establishes a robust and widely applicable benchmark for GFMs. We evaluate the most popular GFMs openly available on this benchmark and analyze their performance across several domains. In particular, we compare these models to supervised baselines [e.g., UNet and vanilla vision-Transformer (ViT)] and assess their effectiveness when faced with limited labeled data. Our findings highlight the limitations of GFMs under different scenarios, showing that they do not consistently outperform supervised models. PANGAEA is designed to be highly extensible, allowing for the seamless inclusion of new datasets, models, and tasks in future research. By releasing the evaluation code and benchmark, we aim to enable other researchers to replicate our experiments and build upon our work, fostering a more principled evaluation protocol for large pretrained geospatial models. The code is available at https://github.com/VMarsocci/pangaea-bench.