1: \begin{abstract}

2:   We study a class of iterative combinatorial auctions which can be

3:   viewed as subgradient descent methods for the problem of pricing

4:   bundles to balance supply and demand. We provide concrete

5:   convergence rates for auctions in this class, bounding the number of

6:   auction rounds needed to reach clearing prices. Our analysis allows

7:   for a variety of pricing schemes, including item, bundle, and

8:   polynomial pricing, and the respective convergence rates confirm

9:   that more expressive pricing schemes come at the cost of slower

10:   convergence. We consider two models of bidder behavior. In the first

11:   model, bidders behave stochastically according to a random utility

12:   model, which includes standard best-response bidding as a special

13:   case. In the second model, bidders can behave arbitrarily (even

14:   adversarially), and meaningful convergence relies on properly

15:   designed activity rules.

16: \end{abstract}