Message passing Graph Neural Networks (GNNs) provide a powerful modeling framework for relational data. However, the expressive power of existing GNNs is upper-bounded by the 1-Weisfeiler-Lehman (1-WL) graph isomorphism test, which means GNNs that are not able to predict node clustering coefficients and shortest path distances, and cannot differentiate between different d-regular graphs. Here we develop a class of message passing GNNs, named Identity-aware Graph Neural Networks (ID-GNNs), with greater expressive power than the 1-WL test. ID-GNN offers a minimal but powerful solution to limitations of existing GNNs. ID-GNN extends existing GNN architectures by inductively considering nodes’ identities during message passing. To embed a given node, ID-GNN first extracts the ego network centered at the node, then conducts rounds of heterogeneous message passing, where different sets of parameters are applied to the center node than to other surrounding nodes in the ego network. We further propose a simplified but faster version of ID-GNN that injects node identity information as augmented node features. Alto- gether, both versions of ID-GNN represent general extensions of message passing GNNs, where experiments show that transforming existing GNNs to ID-GNNs yields on average 40% accuracy improvement on challenging node, edge, and graph property prediction tasks; 3% accuracy improvement on node and graph classification benchmarks; and 15% ROC AUC improvement on real-world link prediction tasks. Additionally, ID-GNNs demonstrate improved or comparable performance over other task-specific graph networks.

Acoustic monitoring is an extremely useful tool for conservationists to track and identify elephants deep within densely wooded environments. However, traditional acoustic audio event detection approaches achieve limited success for two reasons. First, elephant calls are rare in continuous audio data and vary significantly in their duration, frequency, and harmonics. Second, other rare forest noises can easily be mistaken for elephant calls. To tackle class imbalance, we use majority class undersampling and introduce a data augmentation technique that generates additional realistic positive elephant call training examples. Moreover, to handle rare, challenging background sounds, we introduce a novel curriculum driven training strategy that identifies hard to classify background, and uses such sounds to incrementally improve the model. We show that the curriculum approach significantly improves upon our baseline model. Additionally, we introduce the limit case of the curriculum strategy, captured by a new two-stage modeling framework. Our results show that combining these approaches of data augmentation and curriculum-driven learning leads to significant model improvement, achieving a test F1 score of 0.74 compared to 0.45 for the baseline model.

Artificial intelligence is increasingly used in ecological contexts to monitor animal and insect populations. Species of interest are those in danger of extinction, and those that play pivotal roles in agriculture. Noticing population declines or geographical shifts early enough for intervention can prevent local famine and disruption to the global food chain. Traditionally, data are collected in the field using human labor or sensors. Applicable classification models then analyze the data on central servers. The most expensive, and sometimes dangerous part of the remote sensing solution is the human labor of visiting the sensors, retrieving data, and changing batteries. Constantly sending all readings by radio is expensive in power. Instead, having AI in the sensors process readings, and only transmitting results could lead to an indefinitely autonomous, renewably powered solution. We implemented an elephant vocalization detector on a small processor board, and demonstrate that such a device can be operated at low enough power levels with considerable freedom of choice among AI technologies. We achieved a mean of 1.6W, in the best case staying within 75% of memory limits. Measurements covered three inference models, two batch sizes, and two floating point word width settings.

Real-world problems are often not fully characterized by a single optimal solution, as they frequently involve multiple competing objectives; it is therefore important to identify the so-called Pareto frontier, which captures solution trade-offs. We propose a fully polynomial-time approximation scheme based on Dynamic Programming (DP) for computing a polynomially succinct curve that approximates the Pareto frontier to within an arbitrarily small epsilon > 0 on tree-structured networks. Given a set of objectives, our approximation scheme runs in time polynomial in the size of the instance and 1/epsilon. We also propose a Mixed Integer Programming (MIP) scheme to approximate the Pareto frontier. The DP and MIP Pareto frontier approaches have complementary strengths and are surprisingly effective. We provide empirical results showing that our methods outperform other approaches in efficiency and accuracy. Our work is motivated by a problem in computational sustainability concerning the proliferation of hydropower dams throughout the Amazon basin. Our goal is to support decision-makers in evaluating impacted ecosystem services on the full scale of the Amazon basin. Our work is general and can be applied to approximate the Pareto frontier of a variety of multiobjective problems on tree-structured networks.

We provide an exact and approximation algorithm based on Dynamic Programming and an approximation algorithm based on Mixed Integer Programming for optimizing for the so-called dendritic connectivity on tree-structured networks in a multi-objective setting. Dendritic connectivity describes the degree of connectedness of a network. We consider different variants of dendritic connectivity to capture both network connectivity with respect to long and short-to-middle distances. Our work is motivated by a problem in computational sustainability concerning the evaluation of trade-offs in ecosystem services due to the proliferation of hydropower dams throughout the Amazon basin. In particular, we consider trade-offs between energy production and river connectivity. River fragmentation can dramatically affect fish migrations and other ecosystem services, such as navigation and transportation. In the context of river networks, different variants of dendritic connectivity are important to characterize the movements of different fish species and human populations. Our approaches are general and can be applied to optimizing for dendritic connectivity for a variety of multi-objective problems on tree-structured networks.