Interdependent Scheduling Games

Authors: Andres Abeliuk, Haris Aziz, Gerardo Berbeglia, Serge Gaspers, Petr Kalina, Nicholas Mattei, Dominik Peters, Paul Stursberg, Pascal Van Hentenryck, Toby Walsh

IJCAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We implemented the ILP and solved 1000 randomly generated instances where (a) general rewards are drawn from [50,100] and (b) rewards are uniform. The dependency graphs are generated by first randomly permuting the list of all services; then for each service i, drawing a random number of child services c 2 {0, 1, 2} and adding edge (i, i + c) with probability 0.5. Increasing the number/likelihood of dependencies by increasing the potential number of children or increasing the connection probability significantly increases runtime. Figure 1 shows the results for different parameters using Gurobi 6.5 on a computer equipped with an 2.0 GHz Intel Xeon E5405 CPU with 4 GB of RAM.
Researcher Affiliation Collaboration Andres Abeliuk Data61/NICTA andres.abeliuk@data61.csiro.au Haris Aziz Data61/NICTA and UNSW haris.aziz@data61.csiro.au Gerardo Berbeglia University of Melbourne g.berbeglia@mbs.edu Serge Gaspers UNSW and Data61/NICTA sergeg@cse.unsw.edu.au Petr Kalina Czech Technical University petr.kalina@fel.cvut.cz Nicholas Mattei Data61/NICTA and UNSW nicholas.mattei@data61.csiro.au Dominik Peters University of Oxford dominik.peters@cs.ox.ac.uk Paul Stursberg Technische Universit at M unchen paul.stursberg@ma.tum.de Pascal Van Hentenryck University of Michigan pvanhent@umich.edu Toby Walsh UNSW and Data61/NICTA tw@cse.unsw.edu.au
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide concrete access to source code for the methodology described.
Open Datasets No The paper states: 'We implemented the ILP and solved 1000 randomly generated instances'. This indicates custom generated data, not a publicly available dataset with specific access information.
Dataset Splits No The paper mentions 'randomly generated instances' but does not provide specific dataset split information (like percentages or counts for training, validation, or test sets).
Hardware Specification Yes Figure 1 shows the results for different parameters using Gurobi 6.5 on a computer equipped with an 2.0 GHz Intel Xeon E5405 CPU with 4 GB of RAM.
Software Dependencies Yes Figure 1 shows the results for different parameters using Gurobi 6.5 on a computer equipped with an 2.0 GHz Intel Xeon E5405 CPU with 4 GB of RAM.
Experiment Setup Yes We implemented the ILP and solved 1000 randomly generated instances where (a) general rewards are drawn from [50,100] and (b) rewards are uniform. The dependency graphs are generated by first randomly permuting the list of all services; then for each service i, drawing a random number of child services c 2 {0, 1, 2} and adding edge (i, i + c) with probability 0.5. Increasing the number/likelihood of dependencies by increasing the potential number of children or increasing the connection probability significantly increases runtime.