Answering Hindsight Queries with Lifted Dynamic Junction Trees
This work provides an incremental improvement for researchers in probabilistic AI by enabling more efficient hindsight queries in relational temporal models.
The authors tackled the problem of answering hindsight queries in probabilistic relational temporal models by extending the lifted dynamic junction tree algorithm to include an efficient backward pass for smoothing inference, resulting in faster query answering compared to static methods.
The lifted dynamic junction tree algorithm (LDJT) efficiently answers filtering and prediction queries for probabilistic relational temporal models by building and then reusing a first-order cluster representation of a knowledge base for multiple queries and time steps. We extend LDJT to (i) solve the smoothing inference problem to answer hindsight queries by introducing an efficient backward pass and (ii) discuss different options to instantiate a first-order cluster representation during a backward pass. Further, our relational forward backward algorithm makes hindsight queries to the very beginning feasible. LDJT answers multiple temporal queries faster than the static lifted junction tree algorithm on an unrolled model, which performs smoothing during message passing.