“A probabilistic population code based on neural samples” Shivkumar, et al. NIPS 2018
This paper explores an interesting question about the brain – how do sensory neurons represent probability distributions? Do they represent values over parametric models (like models of orientation in V1), or do they represent direct samples of a probability model of the world?
If you want some background on the debate, the paper hinges on the following:
- “Demixing odors – fast inference in olfaction” Grabska-Barwinska, et al. NIPS 2013
- “Bayesian inference with probabilistic population codes” Ma, et al. Nature Neuroscience 2006
Food provided – if you want pizza or salad, please RSVP with Heide at firstname.lastname@example.org .
The Systems Journal Club has spirited discussions of recent systems neuroscience papers, covering sensory, motor and cognitive issues. We meet at Noon on Fridays in the 3rd floor conference room of East McDonnell. If you have any questions, contact Larry Snyder (WashU Dept. of Neuroscience).