We ( @zeniabaek.bsky.social @moicoll.bsky.social and @asgerhobolth.bsky.social ) present a new cool way to visualize the optimal trade off for hmm decoding called Artemis plots! arxiv.org/pdf/2504.15156 www.biorxiv.org/content/10.1... - ThreadSky

lauritsskov.bsky.social • 15 days ago

We ( @zeniabaek.bsky.social @moicoll.bsky.social and @asgerhobolth.bsky.social ) present a new cool way to visualize the optimal trade off for hmm decoding called Artemis plots!
https://arxiv.org/pdf/2504.15156
https://www.biorxiv.org/content/10.1101/2025.04.22.649993v1

Comments

lauritsskov.bsky.social•15 days ago

And thanks to Lidia Gonfaus for making this beautiful illustration! :)

lauritsskov.bsky.social•15 days ago

This part of the manuscript has to do with finding the best decoded path for a hidden markov model so let's talk the two most used: Viterbi and posterior decoding

lauritsskov.bsky.social•15 days ago

Viterbi decoding maximixes the global probability of the hidden state path. While leading to high accuracy this decoding tends to not switch states as often as it should and many short segments will be lost.

lauritsskov.bsky.social•15 days ago

Posterior decoding maximixes the local probability - that is the pointwise accuracy. This leads to more short fragments being recovered but also a higher false positive rate.

lauritsskov.bsky.social•15 days ago

If only there was a way to have your cake and eat it too! Inspired by previous work we propose a weighted mean of posterior and viterbi decoding controlled by a parameter alpha. When alpha = 0 you get posterior decoding and when alpha = 1 you get viterbi. When alpha is intermediate you get a mix.

lauritsskov.bsky.social•15 days ago

The best alpha will depend on the specific emission and transition parameters. Also an alpha of 0.5 does not nescessaraly mark the halfway point between the two methods. There we propose a simulation approach for picking the appropriate alpha for your analyse.

lauritsskov.bsky.social•15 days ago

If you simulate data from your emission and transition parameters you can calculate point wise accuracy/joint posterior probability and sensitivity/specificity plots for each alpha and pick the best one!

Comments

Posting Rules

Reply