Corpus ID: 16107052

Learning in repeated auctions

  title={Learning in repeated auctions},
  author={Thomas Nedelec and Cl{\'e}ment Calauz{\`e}nes and Noureddine El Karoui and Vianney Perchet},
Auction theory historically focused on the question of designing the best way to sell a single item to potential buyers, with the concurrent objectives of maximizing the revenue generated or the welfare created. Those results relied on some prior Bayesian knowledge agents have on each-other and/or on infinite computational power. All those assumptions are no longer satisfied in new markets such as online advertisement: similar items are sold repeatedly, agents are agnostic and try to manipulate… Expand
Multi-Task and Meta-Learning with Sparse Linear Bandits
Motivated by recent developments on metalearning with linear contextual bandit tasks, we study the benefit of feature learning in both the multi-task and meta-learning settings. We focus on the caseExpand
Fast Rate Learning in Stochastic First Price Bidding
This work provides conditions under which the first-price biding utility is locally quadratic around its optimum, and leverages the observation that, on small subintervals, the concentration of the variations of the empirical distribution function may be controlled more accurately than by using the classical Dvoretzky-Kiefer-Wolfowitz inequality. Expand
Provably More Efficient Q-Learning in the One-Sided-Feedback/Full-Feedback Settings
A new Q-learning-based algorithm, Elimination-Based Half-Q-Learning (HQL), that enjoys improved efficiency over existing algorithms for a wide variety of problems in the one-sided-feedback setting and the potential to combine reinforcement learning with richer feedback models is proposed. Expand


Learning to Bid Without Knowing your Value
This work leverages the structure of the utility of the bidder and the partial feedback that bidders typically receive in auctions to provide algorithms with regret rates against the best fixed bid in hindsight that are exponentially faster in convergence in terms of dependence on the action space. Expand
The sample complexity of auctions with side information
This work introduces an alternate model, where bidders are a priori identical, but can be distinguished based (only) on some side information the auctioneer obtains at the time of the auction. Expand
Prior-Independent Optimal Auctions
This work studies the design of optimal prior-independent selling mechanisms, in which the seller faces buyers whose values are drawn from an unknown distribution, and only knows that the distribution belongs to a particular class. Expand
Online learning in repeated auctions
This work considers repeated Vickrey auctions where goods of unknown value are sold sequentially and bidders only learn (potentially noisy) information about a good's value once it is purchased and derives bidding strategies for two models: stochastic and adversarial. Expand
Learning in Repeated Auctions with Budgets: Regret Minimization and Equilibrium
A family of dynamic bidding strategies, referred to as "adaptive pacing" strategies, in which advertisers adjust their bids throughout the campaign according to the sample path of observed expenditures are introduced, which constitute an approximate Nash equilibrium in dynamic strategies. Expand
Thresholding at the monopoly price: an agnostic way to improve bidding strategies in revenue-maximizing auctions
The heart of the approach is to see optimal auctions in practice as a Stackelberg game where the buyer is the leader, as he is the first one to move when the seller is the follower as she has no prior information on the bidder. Expand
Optimal Bidding Strategies and Equilibria in Dynamic Auctions with Budget Constraints
How should agents bid in repeated sequential auctions when they are budget constrained? A motivating example is that of sponsored search auctions, where advertisers bid in a sequence of generalizedExpand
Efficient Regret Bounds for Online Bid Optimisation in Budget-Limited Sponsored Search Auctions
This paper studies the problem of an advertising agent who needs to intelligently distribute her budget across a sequence of online keyword bidding auctions, and shows that although GPL achieves the best performance on average (up to 90% of the optimal solution), its long running time may limit its suitability in practice. Expand
Dynamic Auctions with Bank Accounts
A restricted family of dynamic auctions that can be implemented in an online fashion and without too much commitment from the seller ahead of time are explored, and even with these restrictions, can achieve significantly better revenue and social welfare compared to separate Myerson auctions. Expand
The sample complexity of revenue maximization
It is shown that the only way to achieve a sufficiently good constant approximation of the optimal revenue is through a detailed understanding of bidders' valuation distributions, and introduces α-strongly regular distributions, which interpolate between the well-studied classes of regular and MHR distributions. Expand