Probabilistic reasoning in intelligent systems: networks of plausible inference
Probabilistic reasoning in intelligent systems: networks of plausible inference
Neural decoding with hierarchical generative models
Neural Computation
Hi-index | 0.01 |
We introduce a new approach to neural encoding and decoding which makes use of sparse regression and Markov random fields. We show that interesting response functions were estimated from neuroimaging data acquired while a subject was watching checkerboard patterns and geometrical figures. Furthermore, we demonstrate that reconstructions of the original stimuli can be generated by loopy belief propagation in a Markov random field.