To prevent future outbreaks of COVID-19, Australia is pursuing a mass-vaccination approach in which a targeted group of the population comprising healthcare workers, aged-care residents and other individuals at increased risk of exposure will receive a highly effective priority vaccine. The rest of the population will instead have access to a less effective vaccine. We apply a large-scale agent-based model of COVID-19 in Australia to investigate the possible implications of this hybrid approach to mass-vaccination. The model is calibrated to recent epidemiological and demographic data available in Australia, and accounts for several components of vaccine efficacy. Within a feasible range of vaccine efficacy values, our model supports the assertion that complete herd immunity due to vaccination is not likely in the Australian context. For realistic scenarios in which herd immunity is not achieved, we simulate the effects of mass-vaccination on epidemic growth rate, and investigate the requirements of lockdown measures applied to curb subsequent outbreaks. In our simulations, Australia's vaccination strategy can feasibly reduce required lockdown intensity and initial epidemic growth rate by 43\% and 52\%, respectively. The severity of epidemics, as measured by the peak number of daily new cases, decreases by up to two orders of magnitude under plausible mass-vaccination and lockdown strategies. The study presents a strong argument for a large-scale vaccination campaign, which would significantly reduce the intensity of non-pharmaceutical interventions in Australia and curb future outbreaks.

Inferring linear dependence between time series is central to our understanding of natural and artificial systems. Unfortunately, the hypothesis tests that are used to determine statistically significant directed or multivariate relationships from time-series data often yield spurious associations (Type I errors) or omit causal relationships (Type II errors). This is due to the autocorrelation present in the analyzed time series—a property that is ubiquitous across diverse applications, from brain dynamics to climate change. Here we show that, for limited data, this issue cannot be mediated by fitting a time-series model alone (e.g., in Granger causality or prewhitening approaches), and instead that the degrees of freedom in statistical tests should be altered to account for the effective sample size induced by cross-correlations in the observations. This insight enabled us to derive modified hypothesis tests for any multivariate correlation-based measures of linear dependence between covariance-stationary time series, including Granger causality and mutual information with Gaussian marginals. We use both numerical simulations (generated by autoregressive models and digital filtering) as well as recorded fMRI-neuroimaging data to show that our tests are unbiased for a variety of stationary time series. Our experiments demonstrate that the commonly used F- and χ2-tests can induce significant false-positive rates of up to 100% for both measures, with and without prewhitening of the signals. These findings suggest that many dependencies reported in the scientific literature may have been, and may continue to be, spuriously reported or missed if modified hypothesis tests are not used when analyzing time series.

There is a continuing debate on relative benefits of various mitigation and suppression strategies aimed to control the spread of COVID-19. Here we report the results of agent-based modelling using a fine-grained computational simulation of the ongoing COVID-19 pandemic in Australia. This model is calibrated to match key characteristics of COVID-19 transmission. An important calibration outcome is the age-dependent fraction of symptomatic cases, with this fraction for children found to be one-fifth of such fraction for adults. We apply the model to compare several intervention strategies, including restrictions on international air travel, case isolation, home quarantine, social distancing with varying levels of compliance, and school closures. School closures are not found to bring decisive benefits unless coupled with high level of social distancing compliance. We report several trade-offs, and an important transition across the levels of social distancing compliance, in the range between 70% and 80% levels, with compliance at the 90% level found to control the disease within 13–14 weeks, when coupled with effective case isolation and international travel restrictions.

We examine non-typhoidal Salmonella (S. Typhimurium or STM) epidemics as complex systems, driven by evolution and interactions of diverse microbial strains, and focus on emergence of successful strains. Our findings challenge the established view that seasonal epidemics are associated with random sets of co-circulating STM genotypes. We use high-resolution molecular genotyping data comprising 17,107 STM isolates representing nine consecutive seasonal epidemics in Australia, genotyped by multiple-locus variable-number tandem-repeats analysis (MLVA). From these data, we infer weighted undirected networks based on distances between the MLVA profiles, depicting epidemics as networks of individual bacterial strains. The network analysis demonstrated dichotomy in STM populations which split into two distinct genetic branches, with markedly different prevalences. This distinction revealed the emergence of dominant STM strains defined by their local network topological properties, such as centrality, while correlating the development of new epidemics with global network features, such as small-world propensity.

Understanding animal movements that underpin ecosystem processes is fundamental to ecology. Recent advances in animal tags have increased the ability to remotely locate larger species; however, this technology is not suitable for up to 70% of the world’s bird and mammal species. The most widespread technique for tracking small animals is to manually locate low-power radio transmitters from the ground with handheld equipment. Despite this labor-intensive technique being used for decades, efforts to reduce or automate this process have had limited success. Here, we present an approach for tracking small radio-tagged animals by using an autonomous and lightweight aerial robot. We present experimental results where we used the robot to locate critically endangered swift parrots (Lathamus discolor) within their winter range. The system combines a miniaturized sensor with newly developed estimation algorithms to yield unambiguous bearing- and range-based measurements with associated measures of uncertainty. We incorporated these measurements into Bayesian data fusion and information-based planning algorithms to control the position of the robot as it collected data. We report estimated positions that lie within about 50 meters of the true positions of the birds on average, which are sufficiently accurate for recapture or observation. Further, in comparison with experienced human trackers from locations where the signal was detectable, the robot produced a correct estimate as fast or faster than the human. These results provide validation of robotic systems for wildlife radio telemetry and suggest a way for widespread use as human-assistive or autonomous devices.

In this paper we present AceMod, an agent-based modelling framework for studying influenza epidemics in Australia. The simulator is designed to analyse the spatiotemporal spread of contagion and influenza spatial synchrony across the nation. The individual-based epidemiological model accounts for mobility (worker and student commuting) patterns and human interactions derived from the 2006 Australian census and other national data sources. The high-precision simulation comprises 19.8 million stochastically generated software agents and traces the dynamics of influenza viral infection and transmission at several scales. Using this approach, we are able to synthesise epidemics in Australia with varying outbreak locations and severity. For each scenario, we investigate the spatiotemporal profiles of these epidemics, both qualitatively and quantitatively, via incidence curves, prevalence choropleths, and epidemic synchrony. This analysis exemplifies the nature of influenza pandemics within Australia and facilitates future planning of effective intervention, mitigation and crisis management strategies.

The Kullback–Leibler (KL) divergence is a fundamental measure of information geometry that is used in a variety of contexts in artificial intelligence. We show that, when system dynamics are given by distributed nonlinear systems, this measure can be decomposed as a function of two information-theoretic measures, transfer entropy and stochastic interaction. More specifically, these measures are applicable when selecting a candidate model for a distributed system, where individual subsystems are coupled via latent variables and observed through a filter. We represent this model as a directed acyclic graph (DAG) that characterises the unidirectional coupling between subsystems. Standard approaches to structure learning are not applicable in this framework due to the hidden variables; however, we can exploit the properties of certain dynamical systems to formulate exact methods based on differential topology. We approach the problem by using reconstruction theorems to derive an analytical expression for the KL divergence of a candidate DAG from the observed dataset. Using this result, we present a scoring function based on transfer entropy to be used as a subroutine in a structure learning algorithm. We then demonstrate its use in recovering the structure of coupled Lorenz and Rössler systems.

We propose a decentralized variant of Monte Carlo tree search (MCTS) that is suitable for a variety of tasks in multi-robot active perception. Our algorithm allows each robot to optimize its own actions by maintaining a probability distribution over plans in the joint-action space. Robots periodically communicate a compressed form of their search trees, which are used to update the joint distribution using a distributed optimization approach inspired by variational methods. Our method admits any objective function defined over robot action sequences, assumes intermittent communication, is anytime, and is suitable for online replanning. Our algorithm features a new MCTS tree expansion policy that is designed for our planning scenario. We extend the theoretical analysis of standard MCTS to provide guarantees for convergence rates to the optimal payoff sequence. We evaluate the performance of our method for generalized team orienteering and online active object recognition using real data, and show that it compares favorably to centralized MCTS even with severely degraded communication. These examples demonstrate the suitability of our algorithm for real-world active perception with multiple robots.

We examine salient trends of influenza pandemics in Australia, a rapidly urbanizing nation. To do so, we implement state-of-the-art influenza transmission and progression models within a large-scale stochastic computer simulation, generated using comprehensive Australian census datasets from 2006, 2011, and 2016. Our results offer a simulation-based investigation of a population’s sensitivity to pandemics across multiple historical time points and highlight three notable trends in pandemic patterns over the years: increased peak prevalence, faster spreading rates, and decreasing spatiotemporal bimodality. We attribute these pandemic trends to increases in two key quantities indicative of urbanization: the population fraction residing in major cities and international air traffic. In addition, we identify features of the pandemic’s geographic spread that we attribute to changes in the commuter mobility network. The generic nature of our model and the ubiquity of urbanization trends around the world make it likely for our results to be applicable in other rapidly urbanizing nations.

We develop and apply several novel methods quantifying dynamic multi-agent team interactions. These interactions are detected information-theoretically and captured in two ways: via (i) directed networks (interaction diagrams) representing significant coupled dynamics between pairs of agents, and (ii) state-space plots (coherence diagrams) showing coherent structures in Shannon information dynamics. This model-free analysis relates, on the one hand, the information transfer to responsiveness of the agents and the team, and, on the other hand, the information storage within the team to the team's rigidity and lack of tactical flexibility. The resultant interaction and coherence diagrams reveal implicit interactions, across teams, that may be spatially long-range. The analysis was verified with a statistically significant number of experiments (using simulated football games, produced during RoboCup 2D Simulation League matches), identifying the zones of the most intense competition, the extent and types of interactions, and the correlation between the strength of specific interactions and the results of the matches.

The behaviour of many real-world phenomena can be modelled by nonlinear dynamical systems whereby a latent system state is observed through a filter. We are interested in interacting subsystems of this form, which we model by a set of coupled maps as a synchronous update graph dynamical systems. Specifically, we study the structure learning problem for spatially distributed dynamical systems coupled via a directed acyclic graph. Unlike established structure learning procedures that find locally maximum posterior probabilities of a network structure containing latent variables, our work exploits the properties of dynamical systems to compute globally optimal approximations of these distributions. We arrive at this result by the use of time delay embedding theorems. Taking an information-theoretic perspective, we show that the log-likelihood has an intuitive interpretation in terms of information transfer.

We propose a decentralised variant of Monte Carlo tree search (MCTS) that is suitable for a variety of tasks in multi-robot active perception. Our algorithm allows each robot to optimise its own individual action space by maintaining a probability distribution over plans in the joint-action space. Robots periodically communicate a compressed form of these search trees, which are used to update the locally-stored joint distributions using an optimisation approach inspired by variational methods. Our method admits any objective function defined over robot actions, assumes intermittent communication, and is anytime. We extend the analysis of the standard MCTS for our algorithm and characterise asymptotic convergence under reasonable assumptions. We evaluate the practical performance of our method for generalised team orienteering and active object recognition using real data, and show that it compares favourably to centralised MCTS even with severely degraded communication. These examples support the relevance of our algorithm for real-world active perception with multi-robot systems.

Adversarial patrolling is an algorithmic problem where a robot visits sites within a given area so as to detect the presence of an adversary. We formulate and solve a new variant of this problem where intrusion events occur at discrete locations and are assumed to be clustered in time. Unlike related formulations, we model the behaviour of the adversary using a stochastic point process known as the reactive point process, which naturally models temporally self-exciting events such as pest intrusion and weed growth in agriculture. We present an asymptotically optimal, anytime algorithm based on Monte Carlo tree search that plans the motion of a robot given a separate event detection system in order to regulate event propagation at the sites it visits. We illustrate the behaviour of our algorithm in simulation using several scenarios, and compare its performance to a lawnmower planning algorithm. Our results indicate that our formulation and solution are promising in enabling practical applications and further theoretical extensions.

The application of autonomous robots to efficiently locate small wildlife species has the potential to provide significant ecological insights not previously possible using traditional land-based survey techniques, and a basis for improved conservation policy and management. We present an approach for autonomously localizing radio-tagged wildlife using a small aerial robot. We present a novel two-point phased array antenna system that yields unambiguous bearing measurements and an associated uncertainty measure. Our estimation and information-based planning algorithms incorporate this bearing uncertainty to choose observation points that improve confidence in the location estimate. These algorithms run online in real time and we report experimental results that show successful autonomous localization of stationary radio tags and live radio-tagged birds.

We present several novel methods quantifying dynamic interactions in simulated football games. These interactions are captured in directed networks that represent significant coupled dynamics, detected information-theoretically. The model-free approach measures information dynamics of both pair-wise players’ interactions as well as local tactical contests produced during RoboCup 2D Simulation League games. This analysis involves computation of information transfer and storage, relating the information transfer to responsiveness of the players and the team, and the information storage within the team to the team’s rigidity and lack of tactical flexibility. The resultant directed networks (interaction diagrams) and the measures of responsiveness and rigidity reveal implicit interactions, across teams, that may be delayed and/or long-ranged. The analysis was verified with a number of experiments, identifying the zones of the most intense competition and the extent of interactions.

Robust manipulation with tractability in unstructured environments is a prominent hurdle in robotics. Learning algorithms to control robotic arms have introduced elegant solutions to the complexities faced in such systems. A novel method of Reinforcement Learning (RL), Gaussian Process Dynamic Programming (GPDP), yields promising results for closed-loop control of a low-cost manipulator however research surrounding most RL techniques lack breadth of comparable experiments into the viability of particular learning techniques on equivalent environments. We introduce several model-based learning agents as mechanisms to control a noisy, low-cost robotic system. The agents were tested in a simulated domain for learning closed-loop policies of a simple task with no prior information. Then, the fidelity of the simulations is confirmed by application of GPDP to a physical system.

The RoboCup 2D Simulation League incorporates several challenging features, setting a benchmark for Artificial Intelligence. In this paper we describe some of the ideas and tools used in development of our team, Gliders2013. In doing so, we focus on information dynamics as one of the central mechanisms for tactical analysis. This analysis involves computation of information transfer and storage, relating the information transfer to responsiveness of the players, and the information storage within the team to the team’s rigidity and lack of tactical richness. The proposed approach has been successfully applied to tactical opponent modelling.