Sparse Gaussian Process Approximations Richard Turner University of Cambridge gpss.cc/gpss17/... Wednesday 11am For the full programme and notebooks: gpss.cc/gpss17
Your teaching skills are truly impressive! Thank you for sharing those insightful tips that make these approximations more understandable at their core, especially the link to the factor graphs. Very useful for someone who wishes to connect the dots and gain a comprehensive understanding of why and how these approaches work, rather than just applying their results.
Unfortunately for GPs, it seems for nearly every machine learning-like process the sparse Bernoulli model has been proven as superior, along with the rTop-k algorithm as of late 2020. From my understanding of course.
Your teaching skills are truly impressive! Thank you for sharing those insightful tips that make these approximations more understandable at their core, especially the link to the factor graphs. Very useful for someone who wishes to connect the dots and gain a comprehensive understanding of why and how these approaches work, rather than just applying their results.
Brilliant! It's a very niche topic and this lecture explains the currently available literature in a very intuitive manner.
Amazing explanation. Thanks a lot for uploading. Although I am out of the field, I could grasp it.
Thanks for sharing, very nice explanation!
Turner is so great at teaching, too!
Thanks a lot!
Great explanation, I'm sure it'll get more views as GPs become more mainstream
Unfortunately for GPs, it seems for nearly every machine learning-like process the sparse Bernoulli model has been proven as superior, along with the rTop-k algorithm as of late 2020. From my understanding of course.