国家预印本平台

We investigate the online fair allocation problem with sequentially arriving items under various input models, with the goal of balancing fairness and efficiency. We propose the unconstrained PACE (Pacing According to Current Estimated utility) algorithm, a parameter-free allocation dynamic that requires no prior knowledge of the input while using only integral allocations. PACE attains near-optimal convergence or approximation guarantees under stationary, stochastic-but-nonstationary, and adversarial input types, thereby achieving the first best-of-many-worlds guarantee in online fair allocation. Beyond theoretical bounds, PACE is highly simple, efficient, and decentralized, and is thus likely to perform well on a broad range of real-world inputs. Numerical results support the conclusion that PACE works well under a variety of input models. We find that PACE performs very well on two real-world datasets even under the true temporal arrivals in the data, which are highly nonstationary.

We develop a sparse domination framework adapted to nested families in general measure spaces, possibly including point masses. We show that the Carleson condition is equivalent to a slight generalization of the usual sparse condition, and use this characterization to obtain one- and two-weight estimates for the associated averaging operators, together with converse testing conditions. We apply our general results to integral operators on locally finite trees. Trees come naturally equipped with two nested geometries: sectors and pruned sectors. We prove pointwise sparse domination of natural versions of the harmonic Bergman projector by the corresponding averaging operators, depending on whether the measure on the tree is doubling or not, and show that the distinction is necessary in general. For the positive Bergman projector, we obtain pointwise comparability. As a consequence, we obtain new weighted inequalities and boundedness results for the Bergman projector and a class of Toeplitz-like multipliers.

For a given data distribution $(X_t)_{t \in \mathbb{N}} \sim Q$ i.i.d., we investigate the hypothesis testing problem: $H_0: Q = P_0$ vs. $H_1: Q = P_1$, for two different model probability distributions $P_0$ and $P_1$. In contrast to the standard setting, where analytic densities $p_0$ and $p_1$ are given, here, we consider the density-free setting, where we only have access to i.i.d. simulations $(Z^0_t)_{t \in \mathbb{N}} \sim P_0$ and $(Z^1_t)_{t \in \mathbb{N}} \sim P_1$. For this simulation-based hypothesis testing setting, we construct an e-test martingale, resulting in a sequential test with anytime-valid type-I error guarantees, approximate growth optimality, geometrically decaying type-II error bounds, and asymptotic power one. Most ingredients used in our constructions are variants of well known concepts. The value of this paper lies in the compact presentation of an effective, anytime-valid solution for the density-free simulation-based sequential hypothesis testing case.

Probability forecasts are calibrated when predicted probabilities match empirical outcome frequencies: among events assigned a probability $p$, we'd hope that the fraction of positive outcomes is close to $p$. We study the problem of sequential forecasting of binary outcomes. The classical $O(T^{2/3})$ bound on expected cumulative $\ell_1$-calibration error established by Foster and Vohra stood for over two decades until Dagan et al. reduced the exponent $2/3$ by an unspecified constant. We establish a new two-phase recursive labeling strategy for the sign-preservation-with-reuse game that yields the bound $O(n^αt^β)$ for all choices of space and time. We then sharpen the reduction from upper bounds on sign preservation to calibration by modifying the equivalence of Dagan et al. to use only $O(\log T)$ instances of the sign-preservation-with-reuse game. This lets us establish an explicit bound of $O(T^{0.662942288})$, the first explicit exponent below $2/3$ for sequential calibration, by combining both improvements and choosing explicit feasible parameters.

We show that fictitious play can converge at arbitrarily slow polynomial rates in two-player zero-sum games. For every integer $k \ge 2$, we construct a payoff matrix with $(k+1)^2 - 5$ actions per player for which the duality gap of the empirical strategies decays as $Θ(t^{-1/k})$ after $t$ steps. The family starts from the standard rock-paper-scissors matrix, with each higher-order game constructed recursively from the preceding one. After a prescribed common initial action, every subsequent best response under fictitious play is unique. For $k \ge 3$, these games give counterexamples to Karlin's conjectured $O(t^{-1/2})$ convergence rate, and they extend the recent $Θ(t^{-1/3})$ construction of Wang (2025) to arbitrarily slow polynomial rates.