Tom Heskes

About

Tom Heskes is a leading figure in machine learning and computational neuroscience, whose work bridges theoretical foundations with real-world applications. His research primarily spans self-organizing systems, probabilistic graphical models, and brain-computer interfaces (BCIs). In his seminal 1995 paper, "Self-organization and nonparametric regression," Heskes introduced a powerful framework linking topographic map formation to free energy minimization, deriving an EM-algorithm that remains influential in unsupervised learning. This work, with 9 citations, laid groundwork for understanding neural adaptation. Heskes also made pioneering contributions to BCI technology, as demonstrated in his 2007 study on using high-density magnetoencephalogram (MEG) signals for real-time control of a robot arm. By developing computationally efficient spatial filtering and time-frequency decomposition methods, he enabled practical motor imagery-based control, achieving 8 citations for this impactful demonstration. His work exemplifies how rigorous theoretical insights—from self-organization to probabilistic inference—can drive tangible advances in neural engineering and human-computer interaction.

Research Focus

Key Achievements

Top Papers

1. 1
   Self-organization and nonparametric regression

   9 citations · 1995
2. 2
   ON THE DEVELOPMENT OF A BRAIN-COMPUTER INTERFACE SYSTEM USING HIGH-DENSITY MAGNETOENCEPHALOGRAM SIGNALS FOR REAL-TIME CONTROL OF A ROBOT ARM

   8 citations · 2007

Key Collaborators

Contact & Links