Understanding post-wildfire vegetation loss is critical for developing effective ecological recovery strategies and is often challenging due to the extended time and effort required to capture the evolving ecosystem features. Recent works in this area have not fully explored all the contributing factors, their modalities, and interactions with each other. Furthermore, most research in this domain is limited by a lack of interpretability in predictive modeling, making it less useful in real-world settings. In this work, we propose a novel end-to-end ML pipeline called MVeLMA (Multimodal Vegetation Loss Modeling Architecture) to predict county-wise vegetation loss from fire events. MVeLMA uses a multimodal feature integration pipeline and a stacked ensemble-based architecture to capture different modalities while also incorporating uncertainty estimation through probabilistic modeling. Through comprehensive experiments, we show that our model outperforms several state-of-the-art (SOTA) and baseline models in predicting post-wildfire vegetation loss. Furthermore, we generate vegetation loss confidence maps to identify high-risk counties, thereby helping targeted recovery efforts. The findings of this work have the potential to inform future disaster relief planning, ecological policy development, and wildlife recovery management.

Climate change is intensifying rainfall extremes, making high-resolution precipitation projections crucial for society to better prepare for impacts such as flooding. However, current Global Climate Models (GCMs) operate at spatial resolutions too coarse for localized analyses. To address this limitation, deep learning-based statistical downscaling methods offer promising solutions, providing high-resolution precipitation projections with a moderate computational cost. In this work, we introduce a bias-informed conditional diffusion model for statistical downscaling of precipitation. Specifically, our model leverages a conditional diffusion approach to learn distribution priors from large-scale, high-resolution precipitation datasets. The long-tail distribution of precipitation poses a unique challenge for training diffusion models; to address this, we apply gamma correction during preprocessing. Additionally, to correct biases in the downscaled results, we employ a guided-sampling strategy to enhance bias correction. Our experiments demonstrate that the proposed model achieves highly accurate results in an 8× downscaling setting, outperforming previous deterministic methods. The code and dataset are available at Github.

The prompt estimation of traffic incident impacts (TIIs) plays a crucial role in guiding commuters’ trip planning and enhancing the decision-making resilience of transportation agencies. Despite the strong capability of spatiotemporal modeling, the gap between the TII prediction and the dynamic data mining approaches has not been seamlessly filled. (1) The TII evaluation metrics have never been well-defined, although many criteria for TII exist in research works. (2) Previous attempts heavily rely on predefined road network structures and underscore vital features, leading to inaccurate TII predictions. (3) Predicting the spatiotemporal TII using dynamic road networks is more challenging as it requires extracting both abnormal sub-graph and long-range dependencies due to the large variation of incident clearance time. This research proposes RoadFormer, a novel Road-Anchored Adversarial Dynamic Graph Transformer, for predicting unlimited-range spatiotemporal TIIs. (1) We introduce novel criteria for assessing spatiotemporal TIIs and construct two new benchmark datasets to validate the performance of our methods. (2) RoadFormer leverages a road-anchored spatial transformer and an importance-score temporal transformer to form an encoder-decoder framework. The road-anchored spatial transformer prunes unnecessary edges between nodes with a road-anchored cascade attention mechanism, accurately pinpointing the affected sub-graphs. (3) The importance-score temporal transformer highlights abnormal changes in node features with a score-based adversarial training mechanism, enabling predictions to rely on informative feature changes after the accident occurrence. Extensive experiments on real-world datasets demonstrate that RoadFormer outperforms the state-of the-art methods, especially in capturing spatiotemporal dependency patterns and predicting unlimited-range spatiotemporal TIIs.

Recent research has explored the use of graph neural networks (GNNs) for decentralized control in swarm robotics. However, it has been observed that relying solely on local states is insufficient to imitate a centralized control policy. To address this limitation, previous studies proposed incorporating K-hop delayed states into the computation. While this approach shows promise, it can lead to a lack of consensus among distant flock members and the formation of small localized groups, ultimately resulting in task failure. Our approach is to include the delayed states to build a spatiotemporal GNN model (ST-GNN) by two levels of expansion: spatial expansion and temporal expansion. The spatial expansion utilizes K-hop delayed states to broaden the network while temporal expansion, can effectively predict the trend of swarm behavior, making it more robust against local noise. To validate the effectiveness of our approach, we conducted simulations in two distinct scenarios: free flocking and flocking with a leader. In both scenarios, the simulation results demonstrated that our decentralized ST-GNN approach successfully overcomes the limitations of local controllers. We performed a comprehensive analysis on the effectiveness of spatial expansions and temporal expansions independently. The results clearly demonstrate that both significantly improve overall performance. Furthermore, when combined, they achieve the best performance compared to global solution and delayed states solutions. The performance of ST-GNN underscores its potential as an effective and reliable approach for achieving cohesive flocking behavior while ensuring safety and maintaining desired swarm characteristics.

This work develops a node-edge co-evolution model for attributed graph transformation, where both the node and edge attributes undergo changes due to complex interactions. Due to two fundamental obstacles, learning and approximating attributed graph transformation have not been thoroughly explored: 1) the difficulty of jointly considering four types of atomic interactions including nodes-to-edges, nodes-to-nodes, edges-to-nodes, and edges-to-edges interactions. 2) the difficulty of capturing iterative long-range interactions between nodes and edges. To solve these issues, we offer a novel and scalable equilibrium model, NEC∞, with node-edge message passing and edge-node message passing. Additionally, we propose an efficient optimization algorithm that is based on implicit gradient theorem and includes a theoretical analysis of NEC∞. The effectiveness and efficiency of the proposed model have been demonstrated through extensive experiments on synthetic and real-world data sets.

The vast presence of bots on Twitter requires reliable and accurate bot detection methods that differentiate legitimate bots from malicious ones. Despite the success of those methods, they fail to address the following challenges: (1) the huge size of datasets required to train a model to detect bots, (2) the constant evolution in bot accounts to evade automatic detection leads to scarcity in ground truth real-world datasets, and (3) the complexity in learning representations of a heterogeneous attributed network like Twitter. In this paper, we propose a novel framework, ADNET, to detect anomalies in Twitter-attributed networks using the least amount of labeled data. Specifically, we address the limitations of previous methods by proposing a topology-based active learning framework that uses a deep autoencoder to train the model and is able to handle large graphs better than previous methods. Our experimental results demonstrate that the proposed approach outperforms state-of-the-art methods in detecting anomalous bot accounts and reduces the annotation cost in Twitter attributed networks.