More Like this

1. An agent-based model reveals lost person behavior based on data from wilderness search and rescue

   https://doi.org/10.1038/s41598-022-09502-4  

   Hashimoto, Amanda; Heintzman, Larkin; Koester, Robert; Abaid, Nicole (April 2022, Scientific Reports)

   Abstract Thousands of people are reported lost in the wilderness in the United States every year and locating these missing individuals as rapidly as possible depends on coordinated search and rescue (SAR) operations. As time passes, the search area grows, survival rate decreases, and searchers are faced with an increasingly daunting task of searching large areas in a short amount of time. To optimize the search process, mathematical models of lost person behavior with respect to landscape can be used in conjunction with current SAR practices. In this paper, we introduce an agent-based model of lost person behavior which allows agents to move on known landscapes with behavior defined as independent realizations of a random variable. The behavior random variable selects from a distribution of six known lost person reorientation strategies to simulate the agent’s trajectory. We systematically simulate a range of possible behavior distributions and find a best-fit behavioral profile for a hiker with the International Search and Rescue Incident Database. We validate these results with a leave-one-out analysis. This work represents the first time-discrete model of lost person dynamics validated with data from real SAR incidents and has the potential to improve current methods for wilderness SAR. 

   more » « less
2. Safe reopening of university campuses is possible with COVID-19 vaccination

   https://doi.org/10.1371/journal.pone.0270106  

   Junge, Matthew; Li, Sheng; Samaranayake, Samitha; Zalesak, Matthew (July 2022, PLOS ONE)

   Khudyakov, Yury E (Ed.)

   We construct an agent-based SEIR model to simulate COVID-19 spread at a 16000-student mostly non-residential urban university during the Fall 2021 Semester. We find that mRNA vaccine coverage at 100% combined with weekly screening testing of 25% of the campus population make it possible to safely reopen to in-person instruction. Our simulations exhibit a right-skew for total infections over the semester that becomes more pronounced with less vaccine coverage, less vaccine effectiveness and no additional preventative measures. This suggests that high levels of infection are not exceedingly rare with campus social connections the main transmission route. Finally, we find that if vaccine coverage is 100% and vaccine effectiveness is above 80%, then a safe reopening is possible even without facemask use. This models possible future scenarios with high coverage of additional “booster” doses of COVID-19 vaccines. 

   more » « less
3. A Formal Framework for Knowledge Acquisition: Going beyond Machine Learning

   https://doi.org/10.3390/e24101469  

   Hössjer, Ola; Díaz-Pachón, Daniel Andrés; Rao, J. Sunil (October 2022, Entropy)

   Philosophers frequently define knowledge as justified, true belief. We built a mathematical framework that makes it possible to define learning (increasing number of true beliefs) and knowledge of an agent in precise ways, by phrasing belief in terms of epistemic probabilities, defined from Bayes’ rule. The degree of true belief is quantified by means of active information I+: a comparison between the degree of belief of the agent and a completely ignorant person. Learning has occurred when either the agent’s strength of belief in a true proposition has increased in comparison with the ignorant person (I+>0), or the strength of belief in a false proposition has decreased (I+<0). Knowledge additionally requires that learning occurs for the right reason, and in this context we introduce a framework of parallel worlds that correspond to parameters of a statistical model. This makes it possible to interpret learning as a hypothesis test for such a model, whereas knowledge acquisition additionally requires estimation of a true world parameter. Our framework of learning and knowledge acquisition is a hybrid between frequentism and Bayesianism. It can be generalized to a sequential setting, where information and data are updated over time. The theory is illustrated using examples of coin tossing, historical and future events, replication of studies, and causal inference. It can also be used to pinpoint shortcomings of machine learning, where typically learning rather than knowledge acquisition is in focus. 

   more » « less
4. From Flatland to Jupiter: Searching for Rules of Interaction Across Biological Scales

   https://doi.org/10.1093/icb/icab159  

   Grillo, Claudia A; Holford, Mandë; Walter, Nils G (July 2021, Integrative and Comparative Biology)

   Synopsis In this future-spanning perspective, we examine how an agent-based model could be used to define general rules for interactions across biological systems and evolutionary time. To date, there have been a number of attempts to simulate the emergence of ecological communities using agent-based models of individuals that have evolving traits. Here we speculate whether it is possible to use this computational modeling to simulate self-organizing systems and, importantly, to decipher universal principles that govern biological interactions. This perspective is a thought exercise, meant to extrapolate from current knowledge to how we may make Jupiter-shot leaps to further advance the biosciences in the 21st century. 

   more » « less
5. Priority queue formulation of agent-based bathtub model for network trip flows in the relative space

   https://doi.org/10.1016/j.trc.2024.104765  

   Martínez, Irene; Jin, Wen-Long (November 2024, Transportation Research Part C: Emerging Technologies)

   Agent-based models have been extensively used to simulate the behavior of travelers in transportation systems because they allow for realistic and versatile modeling of interactions. However, traditional agent-based models suffer from high computational costs and rely on tracking physical locations, raising privacy concerns. This paper proposes an efficient formulation for the agent-based bathtub model (AB2M) in the relative space, where each agent’s trajectory is represented by a time series of the remaining distance to its destination. The AB2M can be understood as a microscopic model that tracks individual trips’ initiation, progression, and completion and is an exact numerical solution of the bathtub model for generic (time-dependent) trip distance distributions. The model can be solved for a deterministic set of trips with a given demand pattern (defined by the start time of each trip and its distance), or it can be used to run Monte Carlo simulations to capture the average behavior and variations of stochastic demand patterns. To enhance the computational efficiency, we introduce a priority queue formulation for AB2M, eliminating the need to update trip positions at each time step and allowing us to run large-scale scenarios with millions of individual trips in seconds. We systematically explore the scaling properties of AB2M and discuss the introduction of biases and numerical errors. Finally, we analyze the upper bound of the computational complexity of the AB2M and the benefits of the priority queue formulation and downscaling on the computational cost. The systematic exploration of scaling properties of the modeling of individual agents in the relative space with the AB2M further enhances its applicability to large-scale transportation systems and opens up opportunities for studying travel time reliability, scheduling, and mode choices. 

   more » « less