Abstract
A contention resolution (CR) scheme is a basic algorithmic primitive, that deals with how to allocate items among a random set S of competing players, while maintaining various properties. We consider the most basic setting where a single item must be allocated to some player in S. Here, in a seminal work, Feige and Vondrak (2006) designed a fair CR scheme when the set S is chosen from a product distribution. We explore whether such fair schemes exist for arbitrary non-product distributions on sets of players S, and whether they admit simple algorithmic primitives. Moreover, can we go beyond fair allocation and design such schemes for all possible achievable allocations.
We show that for any arbitrary distribution on sets of players S, and for any achievable allocation, there exist several natural CR schemes that can be succinctly described, are simple to implement and can be efficiently computed to any desired accuracy. We also characterize the space of achievable allocations for any distribution, give algorithms for computing an optimum fair allocation for arbitrary distributions, and describe other natural fair CR schemes for product distributions. These results are based on matrix scaling and various convex programming relaxations.
Part of the work has been done while the author was a postdoctoral fellow at CWI Amsterdam and TU Eindhoven, and was supported by the ERC Consolidator Grant 617951 and the NWO VICI grant 639.023.812.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Define a distribution \(\tilde{\mathcal {P}}\) on \([n+1]\) players, where for \(S\subseteq [n+1]\), we set \(\tilde{p}_S = p_{S \setminus \{n+1\}} \) if \(n+1 \in S \) and \(\tilde{p}_S=0\) otherwise, and \(g_{n+1} = 1-p_\emptyset -\sum _{i \in [n]} g_i\). Then \(\sum _{i \in [n+1]} g_i=1-p_\emptyset \) and whenever the item is assigned to player \(n+1\) in the CR scheme for \(\tilde{\mathcal {P}}\) we do not assign it at all in the scheme for \(\mathcal {P}\).
- 2.
Suppose \(n>3\) and \(S=\{1\}\) or \(S = \{1,\ldots ,n\}\), each with probability 1/2. Then \(q_1 = 1\) and \(q_i=1/2\) for \(i\ge 2\). For a balanced allocation with \(g=\alpha q\), the best g is \((1/(n-1),1/2(n-1),\ldots ,1/2(n-1))\), but here \(\sum _i g_i < 1-p_\emptyset \).
- 3.
In the example above, this allows one to achieve the allocation \((1/2,1/2(n-1),\ldots ,1/2(n-1))\), which is clearly better.
- 4.
It can always be computed efficiently to any desired accuracy \(\varepsilon >0\) using \(\text {poly}(n,1/\varepsilon )\) samples from \(\mathcal {P}\).
References
Adamczyk, M., WÅodarczyk, M.: Random order contention resolution schemes. In: Foundations of Computer Science (FOCS), pp. 790ā801 (2018)
Bansal, N., Srinivasan, A., Svensson, O.: Lift-and-round to improve weighted completion time on unrelated machines. In: Symposium on Theory of Computing, STOC, pp. 156ā167 (2016)
Bansal, N., Sviridenko, M.: The santa claus problem. In: Symposium on Theory of Computing, STOC, pp. 31ā40. ACM (2006)
Bertsekas, D., Gallagher, R.: Data Networks. Prentice Hall, Hoboken (1992)
Boyd, S., Vandenberghe, L.: Convex Optimization. Cambridge University Press, New York (2004)
Charikar, M.: Greedy approximation algorithms for finding dense components in a graph. In: APPROX, pp. 84ā95 (2000)
Charikar, M., Makarychev, K., Makarychev, Y.: Near-optimal algorithms for unique games. In: Symposium on Theory of Computing, STOC, pp. 205ā214 (2006)
Chekuri, C., VondrĆ”k, J., Zenklusen, R.: Submodular function maximization via the multilinear relaxation and contention resolution schemes. SIAM J. Comput. 43(6), 1831ā1879 (2014)
Cohen, M.B., Madry, A., Tsipras, D., Vladu, A.: Matrix scaling and balancing via box constrained newtonās method and interior point methods. In: Symposium on Foundations of Computer Science, FOCS, pp. 902ā913 (2017)
Dresher, M.: Theory and Applications of Games of Strategy. RAND Corporation, Santa Monica (1951)
Feige, U., VondrĆ”k, J.: Approximation algorithms for allocation problems: improving the factor of 1ā1/e. In: Symposium on Foundations of Computer Science, FOCS, pp. 667ā676 (2006)
Feige, U., VondrĆ”k, J.: The submodular welfare problem with demand queries. Theor. Comput. 6(1), 247ā290 (2010)
Feldman, M., Svensson, O., Zenklusen, R.: Online contention resolution schemes. In: Symposium on Discrete Algorithms, SODA, pp. 1014ā1033 (2016)
Hensher, D.A., Rose, J.M., Greene, W.H.: Applied Choice Analysis. Cambridge University Press, Cambridge (2015)
Im, S., Shadloo, M.: Weighted completion time minimization for unrelated machines via iterative fair contention resolution. In: Symposium on Discrete Algorithms, SODA, pp. 2790ā2809 (2020)
Jaynes, E.T.: Information theory and statistical mechanics. Phys. Rev. 106, 620ā630 (1957)
Karmarkar, N., Karp, R.M.: An efficient approximation scheme for the one-dimensional bin-packing problem. In: Foundations of Computer Science, FOCS, pp. 312ā320 (1982)
Kelly, F.P., Maulloo, A.K., Tan, D.K.: Rate control for communication networks: shadow prices, proportional fairness and stability. J. Oper. Res. Soc. 49(3), 237ā252 (1998)
Kohlberg, E.: On the nucleolus of a characteristic function game. SIAM J. Appl. Math. 20(1), 62ā66 (1971)
Lee, E., Singla, S.: Optimal online contention resolution schemes via ex-ante prophet inequalities. In: European Symposium on Algorithms, ESA, volume 112 of LIPIcs, pp. 57:1ā57:14 (2018)
Linial, N., Samorodnitsky, A., Wigderson, A.: A deterministic strongly polynomial algorithm for matrix scaling and approximate permanents. Combinatorica 20(4), 545ā568 (2000)
Luce, R.D.: Individual Choice Behavior: A Theoretical Analysis. Wiley, New York (1959)
Rothblum, U.G., Schneider, H.: Scalings of matrices which have prespecified row sums and column sums via optimization. Linear Algebra Appl. 114, 737ā764 (1989)
Singh, M., Vishnoi, N.K.: Entropy, optimization and counting. In: Symposium on Theory of Computing, STOC, pp. 50ā59 (2014)
Sinkhorn, R.: A relationship between arbitrary positive matrices and doubly stochastic matrices. Ann. Math. Stat. 35(2), 876ā879 (1964)
Train, K.E.: Discrete Choice Methods with Simulation. Cambridge University Press, Cambridge (2009)
Wainwright, M.J., Jordan, M.I.: Graphical models, exponential families, and variational Inference. Found. TrendsĀ® Mach. Learn. 1(1ā2), 1ā305 (2008)
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
Ā© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Bansal, N., Cohen, I.R. (2021). Contention Resolution, Matrix Scaling andĀ Fair Allocation. In: Koenemann, J., Peis, B. (eds) Approximation and Online Algorithms. WAOA 2021. Lecture Notes in Computer Science(), vol 12982. Springer, Cham. https://doi.org/10.1007/978-3-030-92702-8_16
Download citation
DOI: https://doi.org/10.1007/978-3-030-92702-8_16
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-92701-1
Online ISBN: 978-3-030-92702-8
eBook Packages: Computer ScienceComputer Science (R0)