Cansu Sancaktar
Papers
1
Total Citations
57
H-Index
1
About
Cansu Sancaktar is a leading researcher at the intersection of computational neuroscience and artificial intelligence, with a core focus on active inference, embodied cognition, and deep learning. Her most impactful contribution is the development of PixelAI, an end-to-end, pixel-based deep active inference algorithm that bridges the free energy principle with scalable deep convolutional decoders. This pioneering work, published in 2020 and garnering 57 citations, enables artificial agents to directly process raw visual input for body perception and action—a significant departure from traditional handcrafted feature extraction. By grounding high-dimensional sensory data in the variational inference framework of the brain, Sancaktar has advanced the field's understanding of how autonomous systems can learn to perceive and act in complex environments without explicit supervision. Her research not only deepens our grasp of biological intelligence but also provides a robust computational toolkit for building more adaptive, neurally-plausible AI systems. This work stands as a cornerstone for researchers exploring the synergy between neuroscience and machine learning.
Research Focus
Key Achievements
Top Papers
- 1End-to-End Pixel-Based Deep Active Inference for Body Perception and Action57 citations · 2020