Marcel van Gerven
Papers
4
Total Citations
65
H-Index
3
About
Marcel van Gerven is a computational neuroscientist whose research sits at the intersection of artificial intelligence, neuroscience, and robotics. His work is driven by a compelling vision: understanding how the brain processes perception and action, and translating those insights into more capable AI systems. Van Gerven's most recognized contribution is his development of PixelAI, a pixel-based deep active inference algorithm that bridges the free energy principle from neuroscience with deep convolutional architectures, enabling robots and AI agents to learn body perception and action directly from raw visual input. This work, which has accumulated 57 citations since its 2020 publication, exemplifies his talent for turning biological principles into practical computational tools. Beyond robotics, van Gerven has extended his expertise into clinical neuroscience, exploring how AI can advance intracranial EEG research and improve our understanding of neural signals. His ongoing work on neuroscience-inspired perception-action frameworks and continuous control via transition models further demonstrates a sustained commitment to building AI that learns and adapts the way biological systems do. His research offers students and engineers alike a compelling roadmap for creating more intelligent, brain-inspired machines.
Research Focus
Key Achievements
Top Papers
- 1End-to-End Pixel-Based Deep Active Inference for Body Perception and Action57 citations · 2020
- 2How Does Artificial Intelligence Contribute to iEEG Research?3 citations · 2023
- 3
- 4Learning Policies for Continuous Control via Transition Models2 citations · 2023