Document Actions
Matthias Bethge
Bernstein Award for Computational Neuroscience, 2006
Research on Neural Coding and Inference in Early Vision
The agenda of the research project can be summarized by the two basic questions originally posed by Hermann von Helmholtz: “What are the principles that govern how the visual pathways make inferences from the visual image? How do we use image information to compute these perceptual inferences?” A principal difficulty in the understanding of biological vision is the complexity of the inference problems one encounters both at the level of behaviour as well as at the level of neuronal responses. This complexity mostly results from the large number of degrees of freedoms in the sensory input and in the neuronal responses. Using methods of statistical inference and learning theory, as well as signal processing, nonlinear dynamics and optimization theory, the research addresses the problem of perceptual inference from natural images and its neural basis at different levels:
(A) The development of mathematical generative models of natural images and image transformations using unsupervised learning methods. Particular emphasis is placed on quantitative comparisons of the performance of these models.
(B) The performance of psychophysical studies in order to evaluate the relationship between natural image models and perception.
(C) The development of new efficient methods to predict the spike trains of neurons in response to natural stimuli with the goal of inferring the contribution of these neurons to the image processing performed in the early visual system.
| Awardee | Link | ||
| Max-Planck-Institute for Biological Cybernetics |