Search for: All records

Wildlife trafficking is a global phenomenon posing many negative impacts on socio-environmental systems. Scientific exploration of wildlife trafficking trends and the impact of interventions is signifi-cantly encumbered by a suite of data reuse challenges. We describe a novel, open-access data directory on wildlife trafficking and a corresponding visualization tool that can be used to identify data for multiple purposes, such as exploring wildlife trafficking hotspots and convergence points with other crime, discovering key drivers or deterrents of wildlife trafficking, and uncovering structural patterns. Keyword searches, expert elicitation, and peer- reviewed publications were used to search for extant sources used by industry and non-profit organizations, as well as those leveraged to publish academic research articles. The open-access data direc-tory is designed to be a living document and searchable according to multiple measures. The directory can be instrumental in the data- driven analysis of unsustainable illegal wildlife trade, supply chain structure via link prediction models, the value of demand and supply reduction initiatives via multi-item knapsack problems, or trafficking behavior and transportation choices via network inter-diction problems.

Abstract

<p>We describe a novel database on wildlife trafficking that can be used for exploring supply chain coordination via game-theoretic collaboration models, geographic spread of wildlife products trafficked via multi-item knapsack problems, or illicit network interdiction via multi-armed bandit problems.</p>

Other

A publicly available visualization of this dataset is available at: https://public.tableau.com/views/IWTDataDirectory-Gore/Sheet2?:language&#61;en-US&amp;:display_count&#61;n&amp;:origin&#61;viz_share_link

AI for good (AI4G) projects involve developing and applying ar- tificial intelligence (AI) based solutions to further goals in areas such as sustainability, health, humanitarian aid, and social justice. Developing and deploying such solutions must be done in collab- oration with partners who are experts in the domain in question and who already have experience in making progress towards such goals. Based on our experiences, we detail the different aspects of this type of collaboration broken down into four high-level cat- egories: communication, data, modeling, and impact, and distill eleven takeaways to guide such projects in the future. We briefly describe two case studies to illustrate how some of these takeaways were applied in practice during our past collaborations.

Models capturing parameterized random walks on graphs have been widely adopted in wildlife conservation to study species dispersal as a function of landscape features. Learning the probabilistic model empowers ecologists to understand animal responses to conservation strategies. By exploiting the connection between random walks and simple electric networks, we show that learning a random walk model can be reduced to finding the optimal graph Laplacian for a circuit. We propose a moment matching strategy that correlates the model’s hitting and commuting times with those observed empirically. To find the best Laplacian, we propose a neural network capable of back-propagating gradients through the matrix inverse in an end-to-end fashion. We developed a scalable method called CGInv which back-propagates the gradients through a neural network encoding each layer as a conjugate gradient iteration. To demonstrate its effectiveness, we apply our computational framework to applications in landscape connectivity modeling. Our experiments successfully demonstrate that our framework effectively and efficiently recovers the ground-truth configurations.

Motivated by real-world deployment of drones for conservation, this paper advances the state-of-the-art in security games with signaling. The well-known defender-attacker security games framework can help in planning for such strategic deployments of sensors and human patrollers, and warning signals to ward off adversaries. However, we show that defenders can suffer significant losses when ignoring real-world uncertainties despite carefully planned security game strategies with signaling. In fact, defenders may perform worse than forgoing drones completely in this case. We address this shortcoming by proposing a novel game model that integrates signaling and sensor uncertainty; perhaps surprisingly, we show that defenders can still perform well via a signaling strategy that exploits uncertain real-time information. For example, even in the presence of uncertainty, the defender still has an informational advantage in knowing that she has or has not actually detected the attacker; and she can design a signaling scheme to “mislead” the attacker who is uncertain as to whether he has been detected. We provide theoretical results, a novel algorithm, scale-up techniques, and experimental results from simulation based on our ongoing deployment of a conservation drone system in South Africa.