Probabilistic Inference Over Repeated Insertion Models

Authors: Batya Kenig, Lovro Ilijasić, Haoyue Ping, Benny Kimelfeld, Julia Stoyanovich

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We also conduct an experimental study of the algorithm over serial and parallelized implementations. Building upon the relationship between inference with rank distributions and counting linear extensions, we investigate the inference problem when restricted to partial orders that lend themselves to efficient counting of their linear extensions.
Researcher Affiliation Academia Batya Kenig Technion Israel Institute of Technology batyak@cs.technion.ac.il Lovro Ilijasi c Drexel University lovro@drexel.edu Haoyue Ping Drexel University hp354@drexel.edu Benny Kimelfeld Technion Israel Institute of Technology bennyk@cs.technion.ac.il Julia Stoyanovich Drexel University stoyanovich@drexel.edu
Pseudocode Yes An algorithm for computing Pr( | σ, Π) Algorithm RIMDP(RIM(σ, Π)) (Figure 1)
Open Source Code No The paper does not provide any explicit statements about releasing source code for the described methodology or links to a code repository.
Open Datasets No The paper states: 'We evaluated the running time of RIMDP on randomly generated posets that exhibit variability in both size and cover width.' It describes the generation process but does not provide access information (link, DOI, formal citation) for these datasets or any other public dataset.
Dataset Splits No The paper describes how the 'poset workload' was generated, but it does not specify any training, validation, or test dataset splits (e.g., percentages, sample counts, or predefined splits) for reproducibility.
Hardware Specification Yes We implemented RIMDP in Java 8, and executed experiments on an Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz, 512GB of RAM, running 64-bit Ubuntu Linux.
Software Dependencies Yes We implemented RIMDP in Java 8, and executed experiments on an Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz, 512GB of RAM, running 64-bit Ubuntu Linux.
Experiment Setup Yes We generated a poset workload separately for each σ as follows. Let p V and p E be the probabilities that an item (resp. a relation) is added to the poset . First, generate a set of items P items(σ), including each item in P with probability p V . Next, generate a random permutation of the items in P, and denote the resulting subranking by τ. Finally, add each relation σi τ σj to with probability p E. We generated 250 posets for each value of m. For m = 30 items, we set p V [0.3, 0.9] and p E [0.05, 0.3]; for m = 60, p V [0.1, 0.5] and p E [0.1, 0.3]; and for m = 100, p V [0.06, 0.09] and p E [0.1, 0.8]. The parallel implementation of RIMDP takes the number of threads as input, and parallelizes the computation of q[υ, i] for each consistent placement υ at time i.