Kevin P Murphy
Kevin P Murphy
Can you point me to any examples of your existing functionality? We are hoping to submit a paper to COLLAS (https://lifelong-ml.cc/Conferences/2023/call) on March 6th, so would like to try this...
Can you point me to any examples of your existing functionality for boundary-free streams? (We are hoping to submit a paper to COLLAS (https://lifelong-ml.cc/Conferences/2023/call) on March 6th, so would like...
no. but google gemini (internal version) is looking promising...
Hi. Unfortunately this project has been put on ice (although maybe @slinderman will pick it up at some point?). In the meantime, please see chapters 8, 9 and 29 of...
See also https://github.com/probml/JSL/blob/main/jsl/lds/mixture_kalman_filter.py
yes, this is still of interest. You may want to lookinto the SMC support in blackjax, and see if we can pass a dynamax model to it for vanilla PF,...
It's true that for RBPF we can just use a simple discrete proposal, so blackjax may be irrelevant. Here is the KF code: https://github.com/probml/dynamax/blob/main/dynamax/linear_gaussian_ssm/inference.py#L370
What I had in mind is to exploit sparsity to speedup the K^2 computation at each step of forwards-backwards. Currently we just use alpha(t) = A*alpha(t-1), and ignore structure in...
Hi @ghuckins . It would be great if you added your jax implementation of kmeans.