Computational machinery of internal decisions in the mind and machines

Datum

04.06.2026

Zeit

14:30 - 16:00

Sprecher

Prof. Dr. Shervin Safavi

Zugehörigkeit

Computational Machinery of Cognition (CMC lab) TU Dresden

Serie

IMB - Seminar

Sprache

en

Hauptthema

Biologie

Andere Themen

Physik, Mathematik, Informatik, Medizin

Beschreibung

Adaptation is the cornerstone of intelligence, shaping the cognitive hierarchy from basic sensation to complex executive function. Central to this adaptability is the capacity to dynamically integrate internal needs with environmental signals, a process known as internal decision-making. Despite its importance, the precise mechanisms by which biological and artificial neural networks realize these processes remain elusive. In this presentation, I demonstrate that internal decision processes are fundamental to understanding both human and machine behavior. I then explore how the underlying machinery (neural networks) realizes these internal decisions. Lastly, I introduce a novel method that allows for the inference of these internal states purely from the dynamics of both biological and artificial networks. Ultimately, by uncovering the computational "engine" of decision-making, we pave the way for a better understanding of diverse cognitive functions, as well as the mechanisms underlying their dysfunction.

Links

Letztmalig verändert: 05.03.2026, 17:20:20