我们介绍了用于插槽，意图分类和虚拟助手评估的大规模数据集 - 数字亚马逊SLU资源包（SLURP）。大规模包含1M现实，平行，标记为虚拟助手的话语，涵盖51种语言，18个域，60个意图和55个插槽。通过任务专业翻译人员将仅英文slurp数据集定位为29属的50种类型多样性的语言来创建大规模。我们还介绍了XLM-R和MT5上的建模结果，包括精确的匹配精度，意图分类精度和插槽填充F1分数。我们已经公开发布了数据集，建模代码和模型。

The most prevalent notions of fairness in machine learning are statistical definitions: they fix a small collection of high-level, pre-defined groups (such as race or gender), and then ask for approximate parity of some statistic of the classifier (like positive classification rate or false positive rate) across these groups. Constraints of this form are susceptible to (intentional or inadvertent) fairness gerrymandering, in which a classifier appears to be fair on each individual group, but badly violates the fairness constraint on one or more structured subgroups defined over the protected attributes (such as certain combinations of protected attribute values). We propose instead to demand statistical notions of fairness across exponentially (or infinitely) many subgroups, defined by a structured class of functions over the protected attributes. This interpolates between statistical definitions of fairness, and recently proposed individual notions of fairness, but it raises several computational challenges. It is no longer clear how to even check or audit a fixed classifier to see if it satisfies such a strong definition of fairness. We prove that the computational problem of auditing subgroup fairness for both equality of false positive rates and statistical parity is equivalent to the problem of weak agnostic learning -which means it is computationally hard in the worst case, even for simple structured subclasses. However, it also suggests that common heuristics for learning can be applied to successfully solve the auditing problem in practice.We then derive two algorithms that provably converge to the best fair distribution over classifiers in a given class, given access to oracles which can optimally solve the agnostic learning problem. The algorithms are based on a formulation of subgroup fairness as a two-player zero-sum game between a Learner (the primal player) and an Auditor (the dual player). Both algorithms compute an equilibrium of this game. We obtain our first algorithm by simulating play of the game by having Learner play an instance of the no-regret Follow the Perturbed Leader algorithm, and having Auditor play best response. This algorithm provably converges to an approximate Nash equilibrium (and thus to an approximately optimal subgroup-fair distribution over classifiers) in a polynomial number of steps. We obtain our second algorithm by simulating play of the game by having both players play Fictitious Play, which enjoys only provably asymptotic convergence, but has the merit of simplicity and faster per-step computation. We implement the Fictitious Play version using linear regression as a heuristic oracle, and show that we can effectively both audit and learn fair classifiers on real datasets.

Reinforcement Learning (RL) is currently one of the most commonly used techniques for traffic signal control (TSC), which can adaptively adjusted traffic signal phase and duration according to real-time traffic data. However, a fully centralized RL approach is beset with difficulties in a multi-network scenario because of exponential growth in state-action space with increasing intersections. Multi-agent reinforcement learning (MARL) can overcome the high-dimension problem by employing the global control of each local RL agent, but it also brings new challenges, such as the failure of convergence caused by the non-stationary Markov Decision Process (MDP). In this paper, we introduce an off-policy nash deep Q-Network (OPNDQN) algorithm, which mitigates the weakness of both fully centralized and MARL approaches. The OPNDQN algorithm solves the problem that traditional algorithms cannot be used in large state-action space traffic models by utilizing a fictitious game approach at each iteration to find the nash equilibrium among neighboring intersections, from which no intersection has incentive to unilaterally deviate. One of main advantages of OPNDQN is to mitigate the non-stationarity of multi-agent Markov process because it considers the mutual influence among neighboring intersections by sharing their actions. On the other hand, for training a large traffic network, the convergence rate of OPNDQN is higher than that of existing MARL approaches because it does not incorporate all state information of each agent. We conduct an extensive experiments by using Simulation of Urban MObility simulator (SUMO), and show the dominant superiority of OPNDQN over several existing MARL approaches in terms of average queue length, episode training reward and average waiting time.

Mean-field games have been used as a theoretical tool to obtain an approximate Nash equilibrium for symmetric and anonymous $N$-player games in literature. However, limiting applicability, existing theoretical results assume variations of a "population generative model", which allows arbitrary modifications of the population distribution by the learning algorithm. Instead, we show that $N$ agents running policy mirror ascent converge to the Nash equilibrium of the regularized game within $\tilde{\mathcal{O}}(\varepsilon^{-2})$ samples from a single sample trajectory without a population generative model, up to a standard $\mathcal{O}(\frac{1}{\sqrt{N}})$ error due to the mean field. Taking a divergent approach from literature, instead of working with the best-response map we first show that a policy mirror ascent map can be used to construct a contractive operator having the Nash equilibrium as its fixed point. Next, we prove that conditional TD-learning in $N$-agent games can learn value functions within $\tilde{\mathcal{O}}(\varepsilon^{-2})$ time steps. These results allow proving sample complexity guarantees in the oracle-free setting by only relying on a sample path from the $N$ agent simulator. Furthermore, we demonstrate that our methodology allows for independent learning by $N$ agents with finite sample guarantees.

Correlated Equilibrium is a solution concept that is more general than Nash Equilibrium (NE) and can lead to outcomes with better social welfare. However, its natural extension to the sequential setting, the \textit{Extensive Form Correlated Equilibrium} (EFCE), requires a quadratic amount of space to solve, even in restricted settings without randomness in nature. To alleviate these concerns, we apply \textit{subgame resolving}, a technique extremely successful in finding NE in zero-sum games to solving general-sum EFCEs. Subgame resolving refines a correlation plan in an \textit{online} manner: instead of solving for the full game upfront, it only solves for strategies in subgames that are reached in actual play, resulting in significant computational gains. In this paper, we (i) lay out the foundations to quantify the quality of a refined strategy, in terms of the \textit{social welfare} and \textit{exploitability} of correlation plans, (ii) show that EFCEs possess a sufficient amount of independence between subgames to perform resolving efficiently, and (iii) provide two algorithms for resolving, one using linear programming and the other based on regret minimization. Both methods guarantee \textit{safety}, i.e., they will never be counterproductive. Our methods are the first time an online method has been applied to the correlated, general-sum setting.

This paper presents a game-theoretic framework to study the interactions of attack and defense for deep learning-based NextG signal classification. NextG systems such as the one envisioned for a massive number of IoT devices can employ deep neural networks (DNNs) for various tasks such as user equipment identification, physical layer authentication, and detection of incumbent users (such as in the Citizens Broadband Radio Service (CBRS) band). By training another DNN as the surrogate model, an adversary can launch an inference (exploratory) attack to learn the behavior of the victim model, predict successful operation modes (e.g., channel access), and jam them. A defense mechanism can increase the adversary's uncertainty by introducing controlled errors in the victim model's decisions (i.e., poisoning the adversary's training data). This defense is effective against an attack but reduces the performance when there is no attack. The interactions between the defender and the adversary are formulated as a non-cooperative game, where the defender selects the probability of defending or the defense level itself (i.e., the ratio of falsified decisions) and the adversary selects the probability of attacking. The defender's objective is to maximize its reward (e.g., throughput or transmission success ratio), whereas the adversary's objective is to minimize this reward and its attack cost. The Nash equilibrium strategies are determined as operation modes such that no player can unilaterally improve its utility given the other's strategy is fixed. A fictitious play is formulated for each player to play the game repeatedly in response to the empirical frequency of the opponent's actions. The performance in Nash equilibrium is compared to the fixed attack and defense cases, and the resilience of NextG signal classification against attacks is quantified.

This paper presents a game theoretic framework for participation and free-riding in federated learning (FL), and determines the Nash equilibrium strategies when FL is executed over wireless links. To support spectrum sensing for NextG communications, FL is used by clients, namely spectrum sensors with limited training datasets and computation resources, to train a wireless signal classifier while preserving privacy. In FL, a client may be free-riding, i.e., it does not participate in FL model updates, if the computation and transmission cost for FL participation is high, and receives the global model (learned by other clients) without incurring a cost. However, the free-riding behavior may potentially decrease the global accuracy due to lack of contribution to global model learning. This tradeoff leads to a non-cooperative game where each client aims to individually maximize its utility as the difference between the global model accuracy and the cost of FL participation. The Nash equilibrium strategies are derived for free-riding probabilities such that no client can unilaterally increase its utility given the strategies of its opponents remain the same. The free-riding probability increases with the FL participation cost and the number of clients, and a significant optimality gap exists in Nash equilibrium with respect to the joint optimization for all clients. The optimality gap increases with the number of clients and the maximum gap is evaluated as a function of the cost. These results quantify the impact of free-riding on the resilience of FL in NextG networks and indicate operational modes for FL participation.

Finding the mixed Nash equilibria (MNE) of a two-player zero sum continuous game is an important and challenging problem in machine learning. A canonical algorithm to finding the MNE is the noisy gradient descent ascent method which in the infinite particle limit gives rise to the {\em Mean-Field Gradient Descent Ascent} (GDA) dynamics on the space of probability measures. In this paper, we first study the convergence of a two-scale Mean-Field GDA dynamics for finding the MNE of the entropy-regularized objective. More precisely we show that for any fixed positive temperature (or regularization parameter), the two-scale Mean-Field GDA with a {\em finite} scale ratio converges to exponentially to the unique MNE without assuming the convexity or concavity of the interaction potential. The key ingredient of our proof lies in the construction of new Lyapunov functions that dissipate exponentially along the Mean-Field GDA. We further study the simulated annealing of the Mean-Field GDA dynamics. We show that with a temperature schedule that decays logarithmically in time the annealed Mean-Field GDA converges to the MNE of the original unregularized objective function.

In many real-world settings agents engage in strategic interactions with multiple opposing agents who can employ a wide variety of strategies. The standard approach for designing agents for such settings is to compute or approximate a relevant game-theoretic solution concept such as Nash equilibrium and then follow the prescribed strategy. However, such a strategy ignores any observations of opponents' play, which may indicate shortcomings that can be exploited. We present an approach for opponent modeling in multiplayer imperfect-information games where we collect observations of opponents' play through repeated interactions. We run experiments against a wide variety of real opponents and exact Nash equilibrium strategies in three-player Kuhn poker and show that our algorithm significantly outperforms all of the agents, including the exact Nash equilibrium strategies.

Various types of Multi-Agent Reinforcement Learning (MARL) methods have been developed, assuming that agents' policies are based on true states. Recent works have improved the robustness of MARL under uncertainties from the reward, transition probability, or other partners' policies. However, in real-world multi-agent systems, state estimations may be perturbed by sensor measurement noise or even adversaries. Agents' policies trained with only true state information will deviate from optimal solutions when facing adversarial state perturbations during execution. MARL under adversarial state perturbations has limited study. Hence, in this work, we propose a State-Adversarial Markov Game (SAMG) and make the first attempt to study the fundamental properties of MARL under state uncertainties. We prove that the optimal agent policy and the robust Nash equilibrium do not always exist for an SAMG. Instead, we define the solution concept, robust agent policy, of the proposed SAMG under adversarial state perturbations, where agents want to maximize the worst-case expected state value. We then design a gradient descent ascent-based robust MARL algorithm to learn the robust policies for the MARL agents. Our experiments show that adversarial state perturbations decrease agents' rewards for several baselines from the existing literature, while our algorithm outperforms baselines with state perturbations and significantly improves the robustness of the MARL policies under state uncertainties.