Funded Industrial Post-Doc: Generative inference for neuro-timeseries data

[Mike Lawrence]

Hey folks! My company is working on tech to help people recover from stroke faster, and while our current inference methods are performing well and have some neat innovations, they are rather neuroscience-typical in being an ad hoc chain of transforms and I'm positive there's room for more powerful inference through proper generative modelling. I've been working on it on the side but it would be great to add a team member with bandwidth specifically allocated to the project, so we're looking to hire someone with probabilistic programming expertise (particularly TFP, which I'm working to transition to from Stan) to join our team.

Here’s the ad on LinkedIn and Indeed, and feel free to reach me at mi...@axemneuro.com if you have any questions about the position : )

Cheers,

Mike

--
Mike Lawrence, PhD

Co-founder & Lead Scientist

Axem Neurotechnology

axemneuro.com

~ Certainty is (usually) folly ~

[Mike Lawrence]

Minor terminology clarification: my use of the term "generative" in the previous post meant to connote merely data-generating "forward" algorithms written in a PPL and thence belief-updated with real data through Bayes. I belatedly realize that some may have a more ML context for "generative" as being specifically associated with GANs; the position certainly wouldn't expect experience with GANs, just probabilistic programming & inference generally.