Open Access

We interpret solving the multi-vehicle routing problem as a team Markov game with partially observable costs. For a given set of customers to serve, the playing agents (vehicles) have the common goal to determine the team-optimal agent routes with minimal total cost. Each agent thereby observes only its own cost. Our multi-agent reinforcement learning approach, the so-called multi-agent Neural Rewriter, builds on the single-agent Neural Rewriter to solve the problem by iteratively rewriting solutions. Parallel agent action execution and partial observability require new rewriting rules for the game. We propose the introduction of a so-called pool in the system which serves as a collection point for unvisited nodes. It enables agents to act simultaneously and exchange nodes in a conflict-free manner. We realize limited disclosure of agent-specific costs by only sharing them during learning. During inference, each agents acts decentrally, solely based on its own cost. First empirical results on small problem sizes demonstrate that we reach a performance close to the employed OR-Tools benchmark which operates in the perfect cost information setting.

In the context of in silico clinical trials, mechanistic computer models for pathophysiology and pharmacology (here Quantitative Systems Pharmacology models, QSP) can greatly support the decision making for drug candidates and elucidate the (potential) response of patients to existing and novel treatments. These models are built on disease mechanisms and then parametrized using (clinical study) data. Clinical variability among patients is represented by alternative model parameterizations, called virtual patients. Despite the complexity of disease modeling itself, using individual patient data to build these virtual patients is particularly challenging given the high-dimensional, potentially sparse and noisy clinical trial data. In this work, we investigate the applicability of simulation-based inference (SBI), an advanced probabilistic machine learning approach, for virtual patient generation from individual patient data and we develop and evaluate the concept of nearest patient fits (SBI NPF), which further enhances the fitting performance. At the example of rheumatoid arthritis where prediction of treatment response is notoriously difficult, our experiments demonstrate that the SBI approaches can capture large inter-patient variability in clinical data and can compete with standard fitting methods in the field. Moreover, since SBI learns a probability distribution over the virtual patient parametrization, it naturally provides the probability for alternative parametrizations. The learned distributions allow us to generate highly probable alternative virtual patient populations for rheumatoid arthritis, which could potentially enhance the assessment of drug candidates if used for in silico trials.

We develop a weak adversarial approach to solving obstacle problems using neural networks. By employing (generalised) regularised gap functions and their properties we rewrite the obstacle problem (which is an elliptic variational inequality) as a minmax problem, providing a natural formulation amenable to learning. Our approach, in contrast to much of the literature, does not require the elliptic operator to be symmetric. We provide an error analysis for suitable discretisations of the continuous problem, estimating in particular the approximation and statistical errors. Parametrising the solution and test function as neural networks, we apply a modified gradient descent ascent algorithm to treat the problem and conclude the paper with various examples and experiments. Our solution algorithm is in particular able to easily handle obstacle problems that feature biactivity (or lack of strict complementarity), a situation that poses difficulty for traditional numerical methods.

Non-targeted metabolomics holds great promise for advancing precision medicine and biomarker discovery. However, identifying compounds from tandem mass spectra remains a challenging task due to the incomplete nature of spectral reference libraries. Augmenting these libraries with simulated mass spectra can provide the necessary references to resolve unmatched spectra, but generating high-quality data is difficult. In this study, we present FIORA, an open-source graph neural network designed to simulate tandem mass spectra. Our main contribution lies in utilizing the molecular neighborhood of bonds to learn breaking patterns and derive fragment ion probabilities. FIORA not only surpasses state-of-the-art fragmentation algorithms, ICEBERG and CFM-ID, in prediction quality, but also facilitates the prediction of additional features, such as retention time and collision cross section. Utilizing GPU acceleration, FIORA enables rapid validation of putative compound annotations and large-scale expansion of spectral reference libraries with high-quality predictions.