Francisco Santos

David Alcántara, Francisco Criado, Francisco Santos

We show that the shifted Lonely Runner Conjecture (sLRC) holds for 5 runners. We also determine that there are exactly 3 primitive tight instances of the conjecture, only two of which are tight for the non-shifted conjecture (LRC). Our proof is computational, relying on a rephrasing of the sLRC in terms of covering radii of certain zonotopes (Henze and Malikiosis, 2017), and on an upper bound for the (integer) velocities to be checked (Malikiosis, Santos and Schymura, 2024+). As a tool for the proof, we devise an algorithm for bounding the covering radius of rational lattice polytopes, based on constructing dyadic fundamental domains.

Anna Stephens, Francisco Santos, Pang-Ning Tan et al.

Graph neural networks (GNN) are a powerful tool for combining imaging and non-imaging medical information for node classification tasks. Cross-network node classification extends GNN techniques to account for domain drift, allowing for node classification on an unlabeled target network. In this paper we present OTGCN, a powerful, novel approach to cross-network node classification. This approach leans on concepts from graph convolutional networks to harness insights from graph data structures while simultaneously applying strategies rooted in optimal transport to correct for the domain drift that can occur between samples from different data collection sites. This blended approach provides a practical solution for scenarios with many distinct forms of data collected across different locations and equipment. We demonstrate the effectiveness of this approach at classifying Autism Spectrum Disorder subjects using a blend of imaging and non-imaging data.

Yue Deng, Francisco Santos, Pang-Ning Tan et al.

Deep learning based weather forecasting (DLWF) models leverage past weather observations to generate future forecasts, supporting a wide range of downstream tasks, including tropical cyclone (TC) trajectory prediction. In this paper, we investigate their vulnerability to adversarial attacks, where subtle perturbations to the upstream weather forecasts can alter the downstream TC trajectory predictions. Although research on adversarial attacks in DLWF models has grown recently, generating perturbed upstream forecasts that reliably steer downstream output toward attacker-specified trajectories remains a challenge. First, conventional TC detection systems are opaque, non-differentiable black boxes, making standard gradient-based attacks infeasible. Second, the extreme rarity of TC events leads to severe class imbalance problem, making it difficult to develop efficient attack methods that will produce the attacker's target trajectories. Furthermore, maintaining physical consistency in adversarially generated forecasts presents another significant challenge. To overcome these limitations, we propose Cyc-Attack, a novel method that perturbs the upstream forecasts of DLWF models to generate adversarial trajectories. First, we pre-train a differentiable surrogate model to approximate the TC detector's output, enabling the construction of gradient-based attacks. Cyc-Attack also employs skewness-aware loss function with kernel dilation strategy to address the imbalance problem. Finally, a distance-based gradient weighting scheme and regularization are used to constrain the perturbations and eliminate spurious trajectories to ensure the adversarial forecasts are realistic and not easily detectable.