Biodiversity is declining globally at an unprecedented rate. Managers urgently need to allocate limited resources to control pest species where interventions have the highest ecological impact. However, many species are hard to detect, and data collection is often expensive, irregular, and incomplete, thus posing significant challenges for machine learning models that traditionally require large and regular datasets. We present a novel deep learning architecture that estimates the spatiotemporal abundance of hard-to-detect species from sparse, zero-inflated, and irregular data. Our method combines Graph Convolutional Networks (GCNs) to model spatial dependencies across monitoring sites with Recurrent Neural Networks (RNNs) to capture long-range temporal dynamics explicitly addresses the challenges of data sparsity, heterogeneity, and irregular sampling. We apply our model to the Crown-of-Thorns Starfish (COTS) on Australia's Great Barrier Reef, a species with devastating impact on coral reefs and a major target of pest control programs. Our method significantly outperforms baseline approaches and the current resource-intensive approach, manta-tow surveillance, in both accuracy and detectability. Simulations indicate a 20% increase in starfish removal efficiency over a year, enabling more effective coral protection. This work demonstrates how tailored deep learning methods can overcome ecological data limitations and substantially improve conservation outcomes.

Many real-world optimisation problems are defined over both categorical and continuous variables, yet efficient optimisation methods such as Bayesian Optimisation (BO) are ill-equipped to handle such mixed-variable search spaces. The optimisation breadth introduced by categorical variables in the mixed-input setting has seen recent approaches operating on local trust regions, but these methods can be greedy in suboptimal regions of the search space. In this paper, we adopt a holistic view and aim to consolidate optimisation of the categorical and continuous sub-spaces under a single acquisition metric. We develop a tree-based method which retains a global view of the optimisation spaces by identifying regions in the search space with high potential candidates which we call value proposals. Our method uses these proposals to make selections on both the categorical and continuous components of the input. We show that this approach significantly outperforms existing mixed-variable optimisation approaches across several mixed-variable black-box optimisation tasks.

Bayesian approaches developed to solve the optimal design of sequential experiments are mathematically elegant but computationally challenging. Recently, techniques using amortization have been proposed to make these Bayesian approaches practical, by training a parameterized policy that proposes designs efficiently at deployment time. However, these methods may not sufficiently explore the design space, require access to a differentiable probabilistic model and can only optimize over continuous design spaces. Here, we address these limitations by showing that the problem of optimizing policies can be reduced to solving a Markov decision process (MDP). We solve the equivalent MDP with modern deep reinforcement learning techniques. Our experiments show that our approach is also computationally efficient at deployment time and exhibits state-of-the-art performance on both continuous and discrete design spaces, even when the probabilistic model is a black box.

In biodiversity conservation, adaptive management (AM) is the principal tool for decision making under uncertainty. AM problems are planning problems that can be modelled using Mixed Observability MDPs (MOMDPs). MOMDPs tackle decision problems where state variables are completely or partially observable. Unfortunately, MOMDP solutions (policy graphs) are too complex to be interpreted by human decision-makers. Here, we provide algorithms to solve K-N-MOMDPs, where K represents the maximum number of fully observable states and N represents the maximum number of alpha-vectors. Our algorithms calculate compact and more interpretable policy graphs from existing MOMDP models and solutions. We apply these algorithms to two computational sustainability applications: optimal release of bio-control agents to prevent dengue epidemics and conservation of the threatened bird species Gouldian finch. The methods dramatically reduce the number of states and alpha-vectors in MOMDP problems without significantly reducing their quality. The resulting policies have small policy graphs (4-6 nodes) that can be easily interpreted by human decision-makers.

In poor data and urgent decision-making applications, managers need to make decisions without complete knowledge of the system dynamics. In biodiversity conservation, adaptive management (AM) is the principal tool for decision-making under uncertainty. AM can be solved using simplified Mixed Observable Markov Decision Processes called hidden model MDPs (hmMDPs) when the unknown dynamics are assumed stationary. hmMDPs provide optimal policies to AM problems by augmenting the MDP state space with an unobservable state variable representing a finite set of predefined models. A drawback in formalising an AM problem is that experts are often solicited to provide this predefined set of models by specifying the transition matrices. Expert elicitation is a challenging and time-consuming process that is prone to biases, and a key assumption of hmMDPs is that the true transition matrix will be included in the candidate model set. We propose an original approach to build a hmMDP with a universal set of predefined models that is capable of solving any 2-state n-action AM problem. Our approach uses properties of the transition matrices to build the model set and is independent of expert input, removing the potential for expert error in the optimal solution. We provide analytical formulations to derive the minimum set of models to include into an hmMDP to solve any AM problems with 2 states and n actions. We assess our universal AM algorithm on two species conservation case studies from Australia and randomly generated problems.

Inspired by the problem of best managing the invasive mosquito Aedes albopictus across the 17 Torres Straits islands of Australia, we aim at solving a Markov decision process on large Susceptible-Infected-Susceptible (SIS) networks that are highly connected. While dynamic programming approaches can solve sequential decision-making problems on sparsely connected networks, these approaches are intractable for highly connected networks. Inspired by our case study, we focus on problems where the probability of nodes changing state is low and propose two approximate dynamic programming approaches. The first approach is a modified version of value iteration where only those future states that are similar to the current state are accounted for. The second approach models the state space as continuous instead of binary, with an on-line algorithm that takes advantage of Bellman's adapted equation. We evaluate the resulting policies through simulations and provide a priority order to manage the 17 infested Torres Strait islands. Both algorithms show promise, with the continuous state approach being able to scale up to high dimensionality (50 nodes). This work provides a successful example of how AI algorithms can be designed to tackle challenging computational sustainability problems.

In many fields in computational sustainability, applications of POMDPs are inhibited by the complexity of the optimal solution. One way of delivering simple solutions is to represent the policy with a small number of alpha-vectors. We would like to find the best possible policy that can be expressed using a fixed number N of alpha-vectors. We call this the N-POMDP problem. The existing solver alpha-min approximately solves finite-horizon POMDPs with a controllable number of alpha-vectors. However alpha-min is a greedy algorithm without performance guarantees, and it is rather slow. This paper proposes three new algorithms, based on a general approach that we call alpha-min-2. These three algorithms are able to approximately solve N-POMDPs. Alpha-min-2-fast (heuristic) and alpha-min-2-p (with performance guarantees) are designed to complement an existing POMDP solver, while alpha-min-2-solve (heuristic) is a solver itself. Complexity results are provided for each of the algorithms, and they are tested on well-known benchmarks. These new algorithms will help users to interpret solutions to POMDP problems in computational sustainability.

Adaptive management is applied in conservation and natural resource management, and consists of making sequential decisions when the transition matrix is uncertain. Informally described as ’learning by doing’, this approach aims to trade off between decisions that help achieve the objective and decisions that will yield a better knowledge of the true transition matrix. When the true transition matrix is assumed to be an element of a finite set of possible matrices, solving a mixed observability Markov decision process (MOMDP) leads to an optimal trade-off but is very computationally demanding. Under the assumption (common in adaptive management) that the true transition matrix is stationary, we propose a polynomial-time algorithm to find a lower bound of the value function. In the corners of the domain of the value function (belief space), this lower bound is provably equal to the optimal value function. We also show that under further assumptions, it is a linear approximation of the optimal value function in a neighborhood around the corners. We evaluate the benefits of our approach by using it to initialize the solvers MO-SARSOP and Perseus on a novel computational sustainability problem and a recent adaptive management data challenge. Our approach leads to an improved initial value function and translates into significant computational gains for both solvers.

In conservation biology and natural resource management, adaptive management is an iterative process of improving management by reducing uncertainty via monitoring. Adaptive management is the principal tool for conserving endangered species under global change, yet adaptive management problems suffer from a poor suite of solution methods. The common approach used to solve an adaptive management problem is to assume the system state is known and the system dynamics can be one of a set of pre-defined models. The solution method used is unsatisfactory, employing value iteration on a discretized belief MDP which restricts the study to very small problems. We show how to overcome this limitation by modelling an adaptive management problem as a restricted Mixed Observability MDP called hidden model MDP (hmMDP). We demonstrate how to simplify the value function, the backup operator and the belief update computation. We show that, although a simplified case of POMDPs, hm-MDPs are PSPACE-complete in the finite-horizon case. We illustrate the use of this model to manage a population of the threatened Gouldian finch, a bird species endemic to Northern Australia. Our simple modelling approach is an important step towards efficient algorithms for solving adaptive management problems.

Scheduling techniques have been intensively studied by several research communities and have been applied to a wide range of applications in computer and manufacturing environments. Most of the scheduling problems are NP-hard. Therefore, heuristics and approximation algorithms must be used for large problems. Obviously these methods are of interest when they provide near optimal solutions and when computational complexity can be controlled. For this purpose, we have developed a method based on the Hopfield neural network model. This approach permits us to solve in an iterative way a scheduling problem, finding a solution through the minimization of an energy function. An interesting property of this approach is its capacity to trade-off the quality for computation time. Indeed, the convergence speed of the minimization process can be tuned by adapting several parameters that influence the quality of the results. By tuning these parameters, we can build a library of a set of run-time executions (contracts) of the Hopfield minimization process with different characteristics (quality, efficiency). We present two applications exploiting the advantage of having available anytime contract algorithms. The first application illustrates how to build a solution of a one machine scheduling problem within a delay that follows a stochastic distribution. The second application deals with unrelated parallel machine scheduling of non preemptive tasks.