L5-BinocularRivalry

L5-BinocularRivalry - Sampling and the Brain Inference...

Info iconThis preview shows pages 1–7. Sign up to view the full content.

View Full Document Right Arrow Icon
Sampling and the Brain: Inference Control and Driving Nick Jones [email protected]
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
So far What we have covered so far. Inference with chemicals and point process models for chemistry (and neurons). The problem of normalization. Sampling (Rejection, Importance, Metropolis, Gibbs). Encoding uncertain information using probabilistic population coding. Nick Jones Sampling and the Brain: Inference, Control and Driving
Background image of page 2
Today: Binocular rivalry and Gibbs sampling We will now illustrate the preceding ideas with a study and implementation of the following paper: Multistability and Perceptual Inference - Samuel J. Gershman Edward Vul Joshua B. Tenenbaum - Neural Computation 2012 You can read about the basics of binocular rivalry here: http://www.scholarpedia.org/article/Binocular_rivalry
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Plan + Rivalry We’ll cover:
Background image of page 4
Images and Perceptions If we are presented with different images to our left eye x L and to our right eye x R then what image s do we perceive? In practice you appear to see the image from one eye dominating for a while and then flipping to the other eye. Each image has N binary pixels and x i ∈ { 0 , 1 } where i ∈ { L , R } . We suppose that our perception of the n th pixel of the image is a linear combination s n = w n x L n - (1 - w n ) x R n . For simplicity we’ll treat w as a binary vector. Perceiving the image as only that from the left eye would be w = 1 . Nick Jones Sampling and the Brain: Inference, Control and Driving
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Occlusion Left eye x L ; Right eye x R ; Perception s ; s n = w n x L n - (1 - w n ) x R n ; w is a binary vector. For fixed images we might expect a statistical model to yield a posterior distributed about a consensus version of w e.g. w = 1 2 . A fusion not a rivalry .
Background image of page 6
Image of page 7
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

Page1 / 16

L5-BinocularRivalry - Sampling and the Brain Inference...

This preview shows document pages 1 - 7. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online