@Article{DiWi2021,
  author     = {DiPilla, Morgan and Wilson, Mark C.},
  title      = {New algorithms for school choice},
  pages      = {8pp},
  journal    = {},
  abstract   = {Algorithms for the school choice problem have been widely studied. A
small number of algorithms for producing a solution dominate the
literature, for example Serial Dictatorship (SD) and the Deferred
Acceptance (DA) algorithm. One main reason for the dominance of these
algorithms is their good (worst-case) axiomatic behaviour with respect
to notions of Pareto efficiency, justified envy and manipulation.
However if we shift the focus to fairness, social welfare, the
inevitable tradeoffs between incompatible axioms, and average-case
analysis, it is far from clear that these algorithms are the last word.

We introduce a 6-parameter family of 40 inequivalent algorithms derived
from DA, most of which have not appeared (to our knowledge) in the
literature before. The family includes SD and the naive and adaptive
versions of the Boston mechanism, the most commonly used school choice
algorithm other than SD and DA. We investigate agent welfare, order bias
and stability via numerical simulation, under the assumption of truthful
preferences. We find that some of the new algorithms are worthy of much
more study. In particular, when we consider order bias, some of the new
algorithms clearly outperform the classic ones, while losing almost
nothing in stability and utilitarian welfare.}, 
  keywords   = {social choice, assignment},
  url_paper  = {https://markcwilson.site/Research/Outputs/DiWi2021.pdf},
  url_slides = {},
  year       = {2021},
}

{"_id":"Hy7gRMHo8PdyCE9si","bibbaseid":"dipilla-wilson-newalgorithmsforschoolchoice-2021","author_short":["DiPilla, M.","Wilson, M. C."],"bibdata":{"bibtype":"article","type":"article","author":[{"propositions":[],"lastnames":["DiPilla"],"firstnames":["Morgan"],"suffixes":[]},{"propositions":[],"lastnames":["Wilson"],"firstnames":["Mark","C."],"suffixes":[]}],"title":"New algorithms for school choice","pages":"8pp","journal":"","abstract":"Algorithms for the school choice problem have been widely studied. A small number of algorithms for producing a solution dominate the literature, for example Serial Dictatorship (SD) and the Deferred Acceptance (DA) algorithm. One main reason for the dominance of these algorithms is their good (worst-case) axiomatic behaviour with respect to notions of Pareto efficiency, justified envy and manipulation. However if we shift the focus to fairness, social welfare, the inevitable tradeoffs between incompatible axioms, and average-case analysis, it is far from clear that these algorithms are the last word. We introduce a 6-parameter family of 40 inequivalent algorithms derived from DA, most of which have not appeared (to our knowledge) in the literature before. The family includes SD and the naive and adaptive versions of the Boston mechanism, the most commonly used school choice algorithm other than SD and DA. We investigate agent welfare, order bias and stability via numerical simulation, under the assumption of truthful preferences. We find that some of the new algorithms are worthy of much more study. In particular, when we consider order bias, some of the new algorithms clearly outperform the classic ones, while losing almost nothing in stability and utilitarian welfare.","keywords":"social choice, assignment","url_paper":"https://markcwilson.site/Research/Outputs/DiWi2021.pdf","url_slides":"","year":"2021","bibtex":"@Article{DiWi2021,\n author = {DiPilla, Morgan and Wilson, Mark C.},\n title = {New algorithms for school choice},\n pages = {8pp},\n journal = {},\n abstract = {Algorithms for the school choice problem have been widely studied. A\nsmall number of algorithms for producing a solution dominate the\nliterature, for example Serial Dictatorship (SD) and the Deferred\nAcceptance (DA) algorithm. One main reason for the dominance of these\nalgorithms is their good (worst-case) axiomatic behaviour with respect\nto notions of Pareto efficiency, justified envy and manipulation.\nHowever if we shift the focus to fairness, social welfare, the\ninevitable tradeoffs between incompatible axioms, and average-case\nanalysis, it is far from clear that these algorithms are the last word.\n\nWe introduce a 6-parameter family of 40 inequivalent algorithms derived\nfrom DA, most of which have not appeared (to our knowledge) in the\nliterature before. The family includes SD and the naive and adaptive\nversions of the Boston mechanism, the most commonly used school choice\nalgorithm other than SD and DA. We investigate agent welfare, order bias\nand stability via numerical simulation, under the assumption of truthful\npreferences. We find that some of the new algorithms are worthy of much\nmore study. In particular, when we consider order bias, some of the new\nalgorithms clearly outperform the classic ones, while losing almost\nnothing in stability and utilitarian welfare.}, \n keywords = {social choice, assignment},\n url_paper = {https://markcwilson.site/Research/Outputs/DiWi2021.pdf},\n url_slides = {},\n year = {2021},\n}\n\n\n","author_short":["DiPilla, M.","Wilson, M. C."],"key":"DiWi2021","id":"DiWi2021","bibbaseid":"dipilla-wilson-newalgorithmsforschoolchoice-2021","role":"author","urls":{" paper":"https://markcwilson.site/Research/Outputs/DiWi2021.pdf"," slides":"https://drive.google.com/uc?export=download&id=1NEXsxwRAx2CWt43v0y0jyZdP1LdOS_xF"},"keyword":["social choice","assignment"],"metadata":{"authorlinks":{}},"downloads":23},"bibtype":"article","biburl":"https://drive.google.com/uc?export=download&id=1NEXsxwRAx2CWt43v0y0jyZdP1LdOS_xF","dataSources":["yx9ivfGLzvFNsg4LH","QdFdNQZBZSvbPbp7t","o2FBsFrZMS3PxguCG","SjBh38uzaGjRinwq8","M2T285kj8vXfTkGDx","2zimNJgtWEiEM2L3J","TAyBmAZAcnenc4YET","BCqQErP8wgW48bwvt","YjweTJPHHEQ85Pems"],"keywords":["social choice","assignment"],"search_terms":["new","algorithms","school","choice","dipilla","wilson"],"title":"New algorithms for school choice","year":2021,"downloads":23}