• Corpus ID: 220280984

Regularized Online Allocation Problems: Fairness and Beyond

  title={Regularized Online Allocation Problems: Fairness and Beyond},
  author={Santiago R. Balseiro and Haihao Lu and Vahab S. Mirrokni},
Online allocation problems with resource constraints have a rich history in computer science and operations research. In this paper, we introduce the \emph{regularized online allocation problem}, a variant that includes a non-linear regularizer acting on the total resource consumption. In this problem, requests repeatedly arrive over time and, for each request, a decision maker needs to take an action that generates a reward and consumes resources. The objective is to simultaneously maximize… 

Figures and Tables from this paper

Group-level Fairness Maximization in Online Bipartite Matching
  • Will Ma, Pan Xu
  • Computer Science
  • 2020
This paper studies online bipartite matching from the perspective of group-level fairness, and the goal is to balance the number of matches made across different groups of online nodes.
Robust Allocations with Diversity Constraints
It is shown in a formal sense that the Nash Welfare rule that maximizes product of agent values is uniquely positioned to be robust when diversity constraints are introduced, while almost all other natural allocation rules fail this criterion.
Fairness-aware Online Price Discrimination with Nonparametric Demand Models
This work proposes an optimal dynamic pricing policy in terms of regret, which enforces the strict price fairness constraint, and extends the algorithm to a more general notion of fairness, which includes demand fairness as a special case.
Robust Auction Design in the Auto-bidding World
This paper focuses on value maximizing bidders with return on spend constraints and shows that the introduction of reserve prices has a novel impact on the market, by choosing reserve prices appropriately and can improve not only the total revenue but also the total welfare.
The Parity Ray Regularizer for Pacing in Auction Markets
Building on recent work, a novel regularizer is introduced to achieve the frequent case in which advertisers seek to reach a certain distribution of impressions over a target population of users, and is integrated into an online mirror descent scheme attaining the optimal order of sub-linear regret.
Batching and Optimal Multi-stage Bipartite Allocations
This work considers K-stage variants of the classic vertex weighted bipartite b-matching and AdWords problems, and identifies a sequence of "polynomials with decreasing degrees" that can be used as strictly concave regularizers of the optimal matching linear program to form this family of convex-programming based matchings.
Outcome-Driven Dynamic Refugee Assignment with Allocation Balancing
This study proposes two new dynamic assignment algorithms to match refugees and asylum seekers to geographic localities within a host country and finds that the allocation balancing algorithm can achieve near-perfect balance over time with virtually no loss in expected employment compared to the pure employment-maximizing algorithm.
POND: Pessimistic-Optimistic oNline Dispatch
A novel online dispatch algorithm, named POND, standing for Pessimistic-Optimistic oNline Dispatch, which achieves high regret and constraint violation, and experiments show that POND achieves low regret with minimal constraint violations.


Diverse Weighted Bipartite b-Matching
A quadratic programming-based approach to solving a supermodular minimization problem that balances diversity and total weight of the solution and defines the price of diversity, a measure of the efficiency loss due to enforcing diversity, and gives a worst-case theoretical bound.
Convex Optimization: Algorithms and Complexity
This monograph presents the main complexity theorems in convex optimization and their corresponding algorithms and provides a gentle introduction to structural optimization with FISTA, saddle-point mirror prox, Nemirovski's alternative to Nesterov's smoothing, and a concise description of interior point methods.
A Dynamic Near-Optimal Algorithm for Online Linear Programming
This paper proposes a learning-based algorithm that works by dynamically updating a threshold price vector at geometric time intervals, where the dual prices learned from the revealed columns in the previous period are used to determine the sequential decisions in the current period.
Convex optimization theory
  • P. Pardalos
  • Computer Science, Mathematics
    Optim. Methods Softw.
  • 2010
The textbook, Convex Optimization Theory (Athena) by Dimitri Bertsekas, provides a concise, well-organized, and rigorous development of convex analysis and convex optimization theory. Several texts
The Santa Claus problem
This work considers the following problem: The Santa Claus has n presents that he wants to distribute among m kids, each kid has an arbitrary value for each present, and develops an O(log log m/log log log m) approximation algorithm for the restricted assignment case of the problem when p<sub>ij</sub>,0 (i.e. when present j has either value p <sub>j</sub> or 0 for each kid).
On the Efficiency-Fairness Trade-off
The trade-off achieved between efficiency and fairness as one selects different objectives is characterized and several concrete managerial prescriptions for the selection problem are developed based on this trade-offs.
The Best of Many Worlds: Dual Mirror Descent for Online Allocation Problems
This paper considers a data-driven setting in which the reward and resource consumption of each request are generated using an input model that is unknown to the decision maker, and designs a general class of algorithms that attain good performance in various inputs models without knowing which type of input they are facing.
Yield optimization of display advertising with ad exchange
This paper brings to bear techniques from revenue management and stochastic optimization, and derive an efficient policy for making real-time ad allocation decisions, and proves that the policy is asymptotically optimal in terms of an arbitrary (i.e., publisher-defined) trade-off between quality delivered to reservation ads and revenue from the exchange.
Fair Dynamic Rationing
This paper demonstrates that for any set of parameters, a simple adaptive policy of projected proportional allocation achieves the best possible fairness guarantee, ex post as well as ex ante, and demonstrates that this policy provides significant improvement over the class of non-adaptive target-fill-rate policies.
Fair dynamic rationing. EC '21
  • 2021