Viterbi Decoding: δ Tables with MAX, Back-Pointers and the Best Path by Backtracking
This lesson is part of the course
In this lesson
You replace the forward sum by a maximum, store for every state which previous state gave that maximum, read the best last state and walk the back-pointers backwards to the most likely weather path.
Buy the course once for $9.99 to unlock every lesson and lab, the final exam and the certificate – lifetime access, no subscription.
Lessons in this course
- Markov Chains Refresher: Transition Tables, Rows That Sum to 1 and Two-Step Probabilities Locked
- Hidden States & Emissions: The Joint Probability of a Weather Path and a Logbook Locked
- The Forward Algorithm: α Tables Step by Step and the Probability of a Logbook Locked
- Viterbi Decoding: δ Tables with MAX, Back-Pointers and the Best Path by Backtracking Locked
- Log Probabilities: Underflow, LN Sums Instead of Products and Comparing Paths Locked
- Backward & Smoothing: β Tables and the Posterior Probability of Each Day's Weather Locked
- Learning from Counts: Transition and Emission Tables from a Labelled Logbook, with Add-One Smoothing Locked
- Toy Part-of-Speech Tagging: Three Hidden Tags, Viterbi on a Short Sentence, Accuracy and a Posterior Chart Locked