TY - GEN
T1 - The Price is (Probably) Right: Learning Market Equilibria from Samples
AU - Lev, Omer
AU - Patel, Neel
AU - Viswanathan, Vignesh
AU - Zick, Yair
N1 - Funding Information:
Lev, Patel and Zick were supported by the Singapore NRF Research Fellowship #R-252-000-750-733. Patel and Zick were also supported by the AI Singapore Award #AISG-RP-2018-009. Viswanathan was supported by the IITKGP Foundation Award. Most of the work was done while all the authors were at the National University of Singapore. The authors would also like to thank the anonymous reviewers of AAAI 2020, AAMAS 2020 and AAMAS 2021 for their informative comments.
Funding Information:
Lev, Patel and Zick were supported by the Singapore NRF Research Fellowship #R-252-000-750-733. Patel and Zick were also supported by the AI Singapore Award #AISG-RP-2018-009. Viswanathan was
Funding Information:
supported by the IITKGP Foundation Award. Most of the work was done while all the authors were at the National University of Singapore. The authors would also like to thank the anonymous reviewers of AAAI 2020, AAMAS 2020 and AAMAS 2021 for their informative comments.
Publisher Copyright:
© 2021 International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). All rights reserved.
PY - 2021
Y1 - 2021
N2 - Equilibrium computation in markets usually considers settings where
player valuation functions are known. We consider the setting where
player valuations are unknown; using a PAC learning-theoretic framework,
we analyze some classes of common valuation functions, and provide
algorithms which output direct PAC equilibrium allocations, not
estimates based on attempting to learn valuation functions. Since there
exist trivial PAC market outcomes with an unbounded worst-case
efficiency loss, we lower-bound the efficiency of our algorithms. While
the efficiency loss under general distributions is rather high, we show
that in some cases (e.g., unit-demand valuations), it is possible to
find a PAC market equilibrium with significantly better utility.
AB - Equilibrium computation in markets usually considers settings where
player valuation functions are known. We consider the setting where
player valuations are unknown; using a PAC learning-theoretic framework,
we analyze some classes of common valuation functions, and provide
algorithms which output direct PAC equilibrium allocations, not
estimates based on attempting to learn valuation functions. Since there
exist trivial PAC market outcomes with an unbounded worst-case
efficiency loss, we lower-bound the efficiency of our algorithms. While
the efficiency loss under general distributions is rather high, we show
that in some cases (e.g., unit-demand valuations), it is possible to
find a PAC market equilibrium with significantly better utility.
KW - Computer Science - Computer Science and Game Theory
KW - Computer Science - Machine Learning
UR - http://www.scopus.com/inward/record.url?scp=85112342170&partnerID=8YFLogxK
M3 - Conference contribution
T3 - Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS
SP - 755
EP - 763
BT - 20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021
PB - International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS)
T2 - 20th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2021
Y2 - 3 May 2021 through 7 May 2021
ER -