Software

Software development is an essential component of our research. We often prototype our solutions and evaluate them in experimental settings or in simulation. The links below direct you to the personal GitHub repositories of the students or postdocs who led the development.

Selected software

NeuRO Python

Performance profiling of machine learning workloads to predict inference time summary statistics and use the predicted summary statistics for service orchestration subject to inference time constraints.

A. Javeed, V. Fodor and G. Dán, "NeuRO: Inference-time Profiling and Orchestration of ML Applications at the Edge", IEEE INFOCOM, 2026

View on GitHub

BATPAL Python

A method for training robust cooperative multi-agent reinforcement learning policies against an unknown adversary, forming an approximate Bayesian perfect equilibrium.

K. Kazari and G. Dán, "Bayesian Robust Cooperative Multi-Agent Reinforcement Learning Against Unknown Adversaries", ICLR 2026

View on GitHub

Saliutl Python/PyTorch

Robust object detection and classification using CNNs.

M. Byrd Victorica, G. Dán and H. Sandberg, "Saliuitl: Ensemble Salience Guided Recovery of Adversarial Patches Against CNNs", CVPR 2025

View on GitHub

SpaNN Python/PyTorch

Detection of patch attacks on object detection and classification using CNNs.

M. Byrd Victorica, G. Dán and H. Sandberg, "SpaNN: Detecting Multiple Adversarial Patches on CNNs by Spanning Saliency Thresholds", SaTML 2025

View on GitHub