Faster Exact MPE and Constrained Optimization with Deterministic Finite State Automata

Authors: Filippo Bistaffa

IJCAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate FABE on standard benchmark datasets for exact MPE inference (i.e., protein, pedigree, grid) [Kishimoto and Marinescu, 2014; Kishimoto et al., 2015] and WCSP (i.e., spot5, mastermind, iscas89) [Marinescu and Dechter, 2009], comparing it with several state of the art approaches.
Researcher Affiliation Academia Filippo Bistaffa IIIA-CSIC filippo.bistaffa@iiia.csic.es
Pseudocode Yes Algorithm 1 Bucket Elimination [Dechter, 2013] Input: A graphical model M = X, D, F , an ordering d. Output: A max probability (resp. min cost) assignment. 1: Partition functions into buckets according to d. 2: Define ψi as the of bucketi associated with Xi. 3: for p n down to 1 do 4: for ψp and messages h1, h2, . . . , hj in bucketp do 5: hp Xp(ψp Nj i=1 hi). 6: Place hp into the largest index variable in its scope. 7: Assign maximizing (resp. minimizing) values in ordering d, consulting functions in each bucket. 8: return Optimal solution value and assignment.
Open Source Code Yes FABE and SMP are implemented in C++.5 We employ the implementations of RBFAOO and toulbar provided by the authors. 5Code available at https://github.com/filippobistaffa/FABE.
Open Datasets Yes We evaluate FABE on standard benchmark datasets4 for exact MPE inference (i.e., protein, pedigree, grid) [Kishimoto and Marinescu, 2014; Kishimoto et al., 2015] and WCSP (i.e., spot5, mastermind, iscas89) [Marinescu and Dechter, 2009], comparing it with several state of the art approaches. 4Online at: www.ics.uci.edu/ dechter/softwares/benchmarks.
Dataset Splits No The paper mentions using 'standard benchmark datasets' but does not specify the exact split percentages, sample counts, or explicit methodology for training, validation, and test sets.
Hardware Specification Yes All experiments have been run on a cluster whose computing nodes have 2.50GHz CPUs and 384 GBytes of RAM.
Software Dependencies No FABE and SMP are implemented in C++. We employ the implementations of RBFAOO and toulbar provided by the authors. It mentions toulbar [Hurley et al., 2016], CPLEX, and GUROBI, but no specific version numbers for any software dependencies are provided within the paper's text.
Experiment Setup Yes For each instance, we compute d using a weighted MIN-FILL heuristic [Dechter, 2013], and we use the same d for both algorithms. We execute RBFAOO with the parameters detailed in authors previous work [Kishimoto and Marinescu, 2014; Kishimoto et al., 2015], including cache size and i parameter. Every other approach is executed with default parameter values. As for Remark 1, for FABE we consider ϵ=10 10.