Lifted Bayesian Filtering in Multiset Rewriting Systems

Sequential Lifted Bayesian Filtering in Multiset Rewriting Systems

Bayesian Filtering for plan and activity recognition is challenging for scenarios that contain many observation equivalent entities (i. e. entities that produce the same observations). This is due to the combinatorial explosion in the number of hypotheses that need to be tracked. However, this class of problems exhibits a certain symmetry that can be exploited for state space representation and...

LiMa: Sequential Lifted Marginal Filtering on Multiset State Descriptions

ing from Identities (Merge) Abstract State Representation Future work includes research about (effi cient) merging strategies Idea • During the inference, there sometimes are identifying observations that enable to introduce evidence about some entities • Identifying sensors are e.g. ID-card sensors or personal devices, Non-identifying sensors are e.g. presence sensors or light switches • Split...

On Representing Biological Systems through Multiset Rewriting

In the post-genomic era, the most prominent biological problems are detecting, describing and analyzing the informational flows that make a set of molecules a living organism [5]. Genomic and proteomic techniques, in fact, are producing the largest set of biological data available ever, but the problem of detecting and describing how these entities (genes and proteins) interact with each other ...

Multiset rewriting by multiset onstraint solving

Constrained Multiset Rewriting

We investigate model checking of a computation model called Constrained Multiset Rewriting Systems (CMRS). A CMRS operates on configurations which are multisets of monadic predicate symbols, each with an argument ranging over the natural numbers. The transition relation is defined by a finite set of rewriting rules which are conditioned by simple inequalities on variables and constants. This mo...