Harald Burgsteiner
Papers
3
Total Citations
85
H-Index
3
About
Harald Burgsteiner is a researcher whose work sits at the intersection of computational neuroscience and robotics, pioneering the application of spiking neural networks to real-world, physical systems. His key contributions center on using biologically inspired "liquid state machines" to enable movement prediction from raw, real-world visual data—a significant step toward more autonomous and adaptive robotic control. In his seminal 2006 paper, "Movement prediction from real-world images using a liquid state machine" (49 citations), Burgsteiner demonstrated how these networks could process dynamic visual input to forecast motion, a foundational advance for neurorobotics. He further extended this paradigm in "Imitation learning with spiking neural networks and real-world devices" (22 citations), showing that robots could learn complex behaviors by observing human actions, all while operating on energy-efficient, event-driven hardware. His work, including an earlier 2005 study (14 citations), collectively bridges the gap between theoretical neural models and practical, embodied intelligence. Burgsteiner’s research remains influential for its proof-of-concept that spiking networks can handle noisy, real-world data, laying groundwork for future low-power, brain-inspired computing systems.
Research Focus
Key Achievements
Top Papers
- 1Movement prediction from real-world images using a liquid state machine49 citations · 2006
- 2Imitation learning with spiking neural networks and real-world devices22 citations · 2006
- 3Movement Prediction from Real-World Images Using a Liquid State Machine14 citations · 2005