Despite a growing recognition of the importance of insurgent group structure on conflict outcomes, there is very little empirical research thereon. Though this problem is rooted in the inaccessibility of data on militant group structure, insurgents frequently publish large volumes of image data on the internet. In this paper, I develop a new methodology that leverages this abundant but underutilized source of data by automating the creation of a social network graph based on co-appearance in photographs using deep learning. Using a trove of 19,115 obituary images published online by the PKK, a Kurdish militant group in Turkey, I demonstrate that an individual's centrality in the resulting co-appearance network is closely correlated with their rank in the insurgent group.

The use of machine learning (ML) methods for prediction and forecasting has become widespread across the quantitative sciences. However, there are many known methodological pitfalls, including data leakage, in ML-based science. In this paper, we systematically investigate reproducibility issues in ML-based science. We show that data leakage is indeed a widespread problem and has led to severe reproducibility failures. Specifically, through a survey of literature in research communities that adopted ML methods, we find 17 fields where errors have been found, collectively affecting 329 papers and in some cases leading to wildly overoptimistic conclusions. Based on our survey, we present a fine-grained taxonomy of 8 types of leakage that range from textbook errors to open research problems. We argue for fundamental methodological changes to ML-based science so that cases of leakage can be caught before publication. To that end, we propose model info sheets for reporting scientific claims based on ML models that would address all types of leakage identified in our survey. To investigate the impact of reproducibility errors and the efficacy of model info sheets, we undertake a reproducibility study in a field where complex ML models are believed to vastly outperform older statistical models such as Logistic Regression (LR): civil war prediction. We find that all papers claiming the superior performance of complex ML models compared to LR models fail to reproduce due to data leakage, and complex ML models don't perform substantively better than decades-old LR models. While none of these errors could have been caught by reading the papers, model info sheets would enable the detection of leakage in each case.

We present the DLR Planetary Stereo, Solid-State LiDAR, Inertial (S3LI) dataset, recorded on Mt. Etna, Sicily, an environment analogous to the Moon and Mars, using a hand-held sensor suite with attributes suitable for implementation on a space-like mobile rover. The environment is characterized by challenging conditions regarding both the visual and structural appearance: severe visual aliasing poses significant limitations to the ability of visual SLAM systems to perform place recognition, while the absence of outstanding structural details, joined with the limited Field-of-View of the utilized Solid-State LiDAR sensor, challenges traditional LiDAR SLAM for the task of pose estimation using point clouds alone. With this data, that covers more than 4 kilometers of travel on soft volcanic slopes, we aim to: 1) provide a tool to expose limitations of state-of-the-art SLAM systems with respect to environments, which are not present in widely available datasets and 2) motivate the development of novel localization and mapping approaches, that rely efficiently on the complementary capabilities of the two sensors. The dataset is accessible at the following url: https://rmc.dlr.de/s3li_dataset

This study discusses the effect of semi-supervised learning in combination with pretrained language models for data-to-text generation. It is not known whether semi-supervised learning is still helpful when a large-scale language model is also supplemented. This study aims to answer this question by comparing a data-to-text system only supplemented with a language model, to two data-to-text systems that are additionally enriched by a data augmentation or a pseudo-labeling semi-supervised learning approach. Results show that semi-supervised learning results in higher scores on diversity metrics. In terms of output quality, extending the training set of a data-to-text system with a language model using the pseudo-labeling approach did increase text quality scores, but the data augmentation approach yielded similar scores to the system without training set extension. These results indicate that semi-supervised learning approaches can bolster output quality and diversity, even when a language model is also present.

We introduce L-hydra (landmarked hyperbolic distance recovery and approximation), a method for embedding network- or distance-based data into hyperbolic space, which requires only the distance measurements to a few 'landmark nodes'. This landmark heuristic makes L-hydra applicable to large-scale graphs and improves upon previously introduced methods. As a mathematical justification, we show that a point configuration in d-dimensional hyperbolic space can be perfectly recovered (up to isometry) from distance measurements to just d+1 landmarks. We also show that L-hydra solves a two-stage strain-minimization problem, similar to our previous (unlandmarked) method 'hydra'. Testing on real network data, we show that L-hydra is an order of magnitude faster than existing hyperbolic embedding methods and scales linearly in the number of nodes. While the embedding error of L-hydra is higher than the error of existing methods, we introduce an extension, L-hydra+, which outperforms existing methods in both runtime and embedding quality.

Small subgraphs (graphlets) are important features to describe fundamental units of a large network. The calculation of the subgraph frequency distributions has a wide application in multiple domains including biology and engineering. Unfortunately due to the inherent complexity of this task, most of the existing methods are computationally intensive and inefficient. In this work, we propose GNNS, a novel representational learning framework that utilizes graph neural networks to sample subgraphs efficiently for estimating their frequency distribution. Our framework includes an inference model and a generative model that learns hierarchical embeddings of nodes, subgraphs, and graph types. With the learned model and embeddings, subgraphs are sampled in a highly scalable and parallel way and the frequency distribution estimation is then performed based on these sampled subgraphs. Eventually, our methods achieve comparable accuracy and a significant speedup by three orders of magnitude compared to existing methods.

As fishermen have noticed this behaviour, they have used both natural and man-made floating objects, or drifting Fish Aggregating Devices (dFADs), as a tool for finding and catching tropical tunas. The use of dFADs in tuna purse-seine fisheries has gradually increased since the 1980s to the present time, where vessels using dFADs now contribute to 36% of the world's total tropical tuna catch (Davies et al., 2014; Wain et al., 2021; ISSF, 2021). These widespread changes have highlighted the need to better understand the potential ecological effects of dFADs on tuna ecology and the marine environment, in order to ensure adequate management of fish stocks and dFAD usage. Indeed, both the dynamics of how and why tuna associate to dFADs are still poorly understood. Regarding the reasons behind tuna aggregation to dFADs, a number of hypotheses have been suggested (Fréon and Dagorn, 2000; Dempster and Taquet, 2004; Castro et al., 2002). Of these, two have gained traction: the "meeting-point" hypothesis, which considers that dFADs facilitate the encounter between individuals or schools, thus constituting larger schools that could benefit survival rates (Castro et al., 2002); and the "indicator-log" hypothesis, by which tunas may be safeguarding the survival of their eggs, larvae and juvenile stages by using drifting objects as indicators of areas where plankton and food is readily available (Hall et al., 1992). This scenario has led some authors to postulate that man-made dFADs could have detrimental effects on tuna populations by creating a so-called "ecological trap" which would lead tuna to remain associated to dFADs even as these drift into areas that could negatively affect the tuna's behaviour and biology (Marsac et al., 2000; Hallier and Gaertner, 2008). To the best of our knowledge, there is yet no sufficient evidence to either confirm or reject this hypothesis (see Dagorn et al. (2012) and references therein). Given the concerns around the widespread use of dFADs in tuna fisheries today, it is not surprising that a considerable amount of research has been devoted to characterizing the dynamics at play when tunas aggregate to dFADs.

The standard problem setting in cooperative multi-agent settings is self-play (SP), where the goal is to train a team of agents that works well together. However, optimal SP policies commonly contain arbitrary conventions ("handshakes") and are not compatible with other, independently trained agents or humans. This latter desiderata was recently formalized by Hu et al. 2020 as the zero-shot coordination (ZSC) setting and partially addressed with their Other-Play (OP) algorithm, which showed improved ZSC and human-AI performance in the card game Hanabi. OP assumes access to the symmetries of the environment and prevents agents from breaking these in a mutually incompatible way during training. However, as the authors point out, discovering symmetries for a given environment is a computationally hard problem. Instead, we show that through a simple adaption of k-level reasoning (KLR) Costa Gomes et al. 2006, synchronously training all levels, we can obtain competitive ZSC and ad-hoc teamplay performance in Hanabi, including when paired with a human-like proxy bot. We also introduce a new method, synchronous-k-level reasoning with a best response (SyKLRBR), which further improves performance on our synchronous KLR by co-training a best response.

Demystifying the delay propagation mechanisms among multiple airports is fundamental to precise and interpretable delay prediction, which is crucial during decision-making for all aviation industry stakeholders. The principal challenge lies in effectively leveraging the spatiotemporal dependencies and exogenous factors related to the delay propagation. However, previous works only consider limited spatiotemporal patterns with few factors. To promote more comprehensive propagation modeling for delay prediction, we propose SpatioTemporal Propagation Network (STPN), a space-time separable graph convolutional network, which is novel in spatiotemporal dependency capturing. From the aspect of spatial relation modeling, we propose a multi-graph convolution model considering both geographic proximity and airline schedule. From the aspect of temporal dependency capturing, we propose a multi-head self-attentional mechanism that can be learned end-to-end and explicitly reason multiple kinds of temporal dependency of delay time series. We show that the joint spatial and temporal learning models yield a sum of the Kronecker product, which factors the spatiotemporal dependence into the sum of several spatial and temporal adjacency matrices. By this means, STPN allows cross-talk of spatial and temporal factors for modeling delay propagation. Furthermore, a squeeze and excitation module is added to each layer of STPN to boost meaningful spatiotemporal features. To this end, we apply STPN to multi-step ahead arrival and departure delay prediction in large-scale airport networks. To validate the effectiveness of our model, we experiment with two real-world delay datasets, including U.S and China flight delays; and we show that STPN outperforms state-of-the-art methods. In addition, counterfactuals produced by STPN show that it learns explainable delay propagation patterns.

Few-shot sound event detection is the task of detecting sound events, despite having only a few labelled examples of the class of interest. This framework is particularly useful in bioacoustics, where often there is a need to annotate very long recordings but the expert annotator time is limited. This paper presents an overview of the second edition of the few-shot bioacoustic sound event detection task included in the DCASE 2022 challenge. A detailed description of the task objectives, dataset, and baselines is presented, together with the main results obtained and characteristics of the submitted systems. This task received submissions from 15 different teams from which 13 scored higher than the baselines. The highest F-score was of 60% on the evaluation set, which leads to a huge improvement over last year's edition. Highly-performing methods made use of prototypical networks, transductive learning, and addressed the variable length of events from all target classes. Furthermore, by analysing results on each of the subsets we can identify the main difficulties that the systems face, and conclude that few-show bioacoustic sound event detection remains an open challenge.