Download presentation
Presentation is loading. Please wait.
Published byEdward Sims Modified over 5 years ago
1
BP in Practice Message Passing Inference Probabilistic Graphical
Models Message Passing BP in Practice
2
Misconception Revisited
A D B C
3
Nonconvergent BP Run
4
Different Variants of BP
Synchronous BP: all messages are updated in parallel synchronous Time (seconds) 2 4 6 8 10 12 14 # messages converged Ising Grid x 100 11 11 12 12 13 13 21 21 22 22 23 23 31 31 32 32 33 33
5
Different Variants of BP
Asynchronous BP: Messages are updated one at a time asynchronous Time (seconds) 2 4 6 8 10 12 14 # messages converged Ising Grid x 100 asynchronous order 2 synchronous 11 12 13 21 22 23 31 32 33
6
Observations Convergence is a local property:
some messages converge soon others may never converge Synchronous BP converges considerably worse than asynchronous Message passing order makes a difference to extent and rate of convergence
7
Informed Message Scheduling
Tree reparameterization (TRP) Pick a tree and pass messages to calibrate 11 12 13 21 22 23 31 32 33
8
Informed Message Scheduling
Tree reparameterization (TRP) Pick a tree and pass messages to calibrate Residual belief propagation (RBP) Pass messages between two clusters whose beliefs over the sepset disagree the most
9
Smoothing (Damping) Messages
Dampens oscillations in messages
11
Summary To achieve BP convergence, two main tricks
Damping Intelligent message ordering Convergence doesn’t guarantee correctness Bad cases for BP – both convergence & accuracy: Strong potentials pulling in different directions Tight loops Some new algorithms have better convergence: Optimization-based view to inference
12
END END END
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.