Mirko Zaffaroni

About

Mirko Zaffaroni is a researcher at the forefront of generative adversarial networks (GANs) and synthetic data generation, with a particular focus on social interaction modeling. His most influential work, "AA-SGAN: Adversarially Augmented Social GAN with Synthetic Data" (2025), introduces a novel framework that enhances the realism of synthetic social trajectories by integrating adversarial augmentation into the GAN training pipeline. This contribution addresses a critical challenge in human behavior modeling—generating diverse, plausible social interactions that can improve autonomous systems and simulation environments. While still early in its impact, the paper has already garnered 2 citations, signaling growing recognition in the field. Zaffaroni’s research bridges computer vision, machine learning, and social robotics, offering practical tools for generating high-fidelity synthetic datasets that reduce reliance on costly real-world data collection. His work is particularly relevant for applications in autonomous navigation, crowd simulation, and human-robot interaction, where realistic social dynamics are essential. As synthetic data continues to transform AI training, Zaffaroni’s innovations position him as an emerging voice in the next wave of generative modeling research.

Research Focus

Key Achievements

Top Papers

1. 1
   AA-SGAN: Adversarially Augmented Social GAN with Synthetic Data

   2 citations · 2025

Key Collaborators

Contact & Links