Vito Mengers is a roboticist with a background in Computer Engineering and an interest in the computational principles underlying perception, action, and intelligence.

He joined SCIoI as a doctoral researcher in September 2021.

Vito’s work focuses on interconnected recursive estimators, systems that continuously update internal models based on sensory input. Originally applied to robust robotic perception, he has shown that these estimators can also support action selection via gradient descent. This tight integration of perception and action enables agents to both respond quickly to environmental changes and actively shape their perception through behavior.

Pushing beyond robotics, Vito explores these systems as a candidate principle of intelligence. He has demonstrated their ability to model complex phenomena such as human visual attention, for example scanpath generation in natural videos, and collective behavior, such as distributed estimation or coordinated motion.
These systems exhibit rich, robust, and explainable behavior, offering insight into the possible computational mechanisms underlying biological intelligence.

Click here to see Vito’s personal website.