This paper studies a cooperative multi-agent multi-armed stochastic bandit problem where agents operate asynchronously -- agent pull times and rates are unknown, irregular, and heterogeneous -- and face the same instance of a K-armed bandit problem. Agents can share reward information to speed up the learning process at additional communication costs. We propose ODC, an on-demand communication protocol that tailors the communication of each pair of agents based on their empirical pull times. ODC is efficient when the pull times of agents are highly heterogeneous, and its communication complexity depends on the empirical pull times of agents. ODC is a generic protocol that can be integrated into most cooperative bandit algorithms without degrading their performance. We then incorporate ODC into the natural extensions of UCB and AAE algorithms and propose two communication-efficient cooperative algorithms. Our analysis shows that both algorithms are near-optimal in regret.
相關內容
The standard class-incremental continual learning setting assumes a set of tasks seen one after the other in a fixed and predefined order. This is not very realistic in federated learning environments where each client works independently in an asynchronous manner getting data for the different tasks in time-frames and orders totally uncorrelated with the other ones. We introduce a novel federated learning setting (AFCL) where the continual learning of multiple tasks happens at each client with different orderings and in asynchronous time slots. We tackle this novel task using prototype-based learning, a representation loss, fractal pre-training, and a modified aggregation policy. Our approach, called FedSpace, effectively tackles this task as shown by the results on the CIFAR-100 dataset using 3 different federated splits with 50, 100, and 500 clients, respectively. The code and federated splits are available at //github.com/LTTM/FedSpace.
Here we consider the communications tactics appropriate for a group of agents that need to "swarm" together in a highly adversarial environment. Specfically, whilst they need to cooperate by exchanging information with each other about their location and their plans; at the same time they also need to keep such communications to an absolute minimum. This might be due to a need for stealth, or otherwise be relevant to situations where communications are signficantly restricted. Complicating this process is that we assume each agent has (a) no means of passively locating others, (b) it must rely on being updated by reception of appropriate messages; and if no such update messages arrive, (c) then their own beliefs about other agents will gradually become out of date and increasingly inaccurate. Here we use a geometry-free multi-agent model that is capable of allowing for message-based information transfer between agents with different intrinsic connectivities, as would be present in a spatial arrangement of agents. We present agent-centric performance metrics that require only minimal assumptions, and show how simulated outcome distributions, risks, and connectivities depend on the ratio of information gain to loss. We also show that checking for too-long round-trip times can be an effective minimal-information filter for determining which agents to no longer target with messages.
In this article, we propose a reactive task allocation architecture for a multi-agent system for scenarios where the tasks arrive at random times and are grouped into multiple queues. Two stage tasks are considered where every task has a beginning, an intermediate and a final part, typical in pick-and-drop and inspect-and-report scenarios. A centralized auction-based task allocation system is proposed, where an auction system takes into consideration bids submitted by the agents for individual tasks, current length of the queues and the waiting times of the tasks in the queues to decide on a task allocation strategy. The costs associated with these considerations, along with the constraints of having unique mappings between tasks and agents and constraints on the maximum number of agents that can be assigned to a queue, results in a Linear Integer Program (LIP) that is solved using the SCIP solver. For the scenario where the queue lengths are penalized but not the waiting times, we demonstrate that the auction system allocates tasks in a manner that all the queue lengths become constant, which is termed balancing. For the scenarios where both the costs are considered, we qualitatively analyse the effect of the choice of the relative weights on the resulting task allocation and provide guidelines for the choice of the weights. We present simulation results that illustrate the balanced allocation of tasks and validate the analysis for the trade-off between the costs related to queue lengths and task waiting times.
This work proposes a dynamic and adversarial resource allocation problem in a graph environment, which is referred to as the dynamic Defender-Attacker Blotto (dDAB) game. A team of defender robots is tasked to ensure numerical advantage at every node in the graph against a team of attacker robots. The engagement is formulated as a discrete-time dynamic game, where the two teams reallocate their robots in sequence and each robot can move at most one hop at each time step. The game terminates with the attacker's victory if any node has more attacker robots than defender robots. Our goal is to identify the necessary and sufficient number of defender robots to guarantee defense. Through a reachability analysis, we first solve the problem for the case where the attacker team stays as a single group. The results are then generalized to the case where the attacker team can freely split and merge into subteams. Crucially, our analysis indicates that there is no incentive for the attacker team to split, which significantly reduces the search space for the attacker's winning strategies and also enables us to design defender counter-strategies using superposition. We also present an efficient numerical algorithm to identify the necessary and sufficient number of defender robots to defend a given graph. Finally, we present illustrative examples to verify the efficacy of the proposed framework.
We develop a Markovian framework for load balancing where classical algorithms such as Power-of-$d$ are combined with auto-scaling mechanisms, which allow the net service capacity to scale up or down in response to the current load within the same timescale of job dynamics. Our framework is inspired by serverless platforms such as Knative where servers are software functions that can be flexibly instantiated in milliseconds according to scaling rules defined by the users of the serverless platform. The main question is how to design such scaling rules to minimize user-perceived delay performance while guaranteeing low energy consumption. For the first time, we investigate this problem when the auto-scaling and load balancing processes operate \emph{asynchronously}, as in Knative. One advantage induced by asynchronism is that jobs do not necessarily need to wait any time a scale-up decision is taken. In our main result, we find a general condition on the structure of scaling rules able to drive mean-field dynamics to delay and relative energy optimality, i.e., a situation where both the user-perceived delay and the relative energy wastage induced by idle servers vanish in the limit where the network demand grows to infinity in proportion to the nominal service capacity. The identified condition suggests to scale up the current net capacity if and only if the mean demand exceeds the rate at which servers become idle and active. Finally, we propose \emph{Rate-Idle}, i.e., a scaling rule that satisfies our optimality condition, and by means of numerical simulations, we show that it improves delay performance over existing (synchronous) schemes.
Federated Learning (FL) has gained increasing interest in recent years as a distributed on-device learning paradigm. However, multiple challenges remain to be addressed for deploying FL in real-world Internet-of-Things (IoT) networks with hierarchies. Although existing works have proposed various approaches to account data heterogeneity, system heterogeneity, unexpected stragglers and scalibility, none of them provides a systematic solution to address all of the challenges in a hierarchical and unreliable IoT network. In this paper, we propose an asynchronous and hierarchical framework (Async-HFL) for performing FL in a common three-tier IoT network architecture. In response to the largely varied delays, Async-HFL employs asynchronous aggregations at both the gateway and the cloud levels thus avoids long waiting time. To fully unleash the potential of Async-HFL in converging speed under system heterogeneities and stragglers, we design device selection at the gateway level and device-gateway association at the cloud level. Device selection chooses edge devices to trigger local training in real-time while device-gateway association determines the network topology periodically after several cloud epochs, both satisfying bandwidth limitation. We evaluate Async-HFL's convergence speedup using large-scale simulations based on ns-3 and a network topology from NYCMesh. Our results show that Async-HFL converges 1.08-1.31x faster in wall-clock time and saves up to 21.6% total communication cost compared to state-of-the-art asynchronous FL algorithms (with client selection). We further validate Async-HFL on a physical deployment and observe robust convergence under unexpected stragglers.
Classic and deep generalized canonical correlation analysis (GCCA) algorithms seek low-dimensional common representations of data entities from multiple ``views'' (e.g., audio and image) using linear transformations and neural networks, respectively. When the views are acquired and stored at different computing agents (e.g., organizations and edge devices) and data sharing is undesired due to privacy or communication cost considerations, federated learning-based GCCA is well-motivated. In federated learning, the views are kept locally at the agents and only derived, limited information exchange with a central server is allowed. However, applying existing GCCA algorithms onto such federated learning settings may incur prohibitively high communication overhead. This work puts forth a communication-efficient federated learning framework for both linear and deep GCCA under the maximum variance (MAX-VAR) formulation. The overhead issue is addressed by aggressively compressing (via quantization) the exchanging information between the computing agents and a central controller. Compared to the unquantized version, our empirical study shows that the proposed algorithm enjoys a substantial reduction of communication overheads with virtually no loss in accuracy and convergence speed. Rigorous convergence analyses are also presented, which is a nontrivial effort. Generic federated optimization results do not cover the special problem structure of GCCA. Our result shows that the proposed algorithms for both linear and deep GCCA converge to critical points at a sublinear rate, even under heavy quantization and stochastic approximations. In addition, in the linear MAX-VAR case, the quantized algorithm approaches a global optimum in a geometric rate under reasonable conditions. Synthetic and real-data experiments are used to showcase the effectiveness of the proposed approach.
The stochastic approximation (SA) algorithm is a widely used probabilistic method for finding a zero or a fixed point of a vector-valued funtion, when only noisy measurements of the function are available. In the literature to date, one makes a distinction between ``synchronous'' updating, whereby every component of the current guess is updated at each time, and ``asynchronous'' updating, whereby only one component is updated. In this paper, we study an intermediate situation that we call ``batch asynchronous stochastic approximation'' (BASA), in which, at each time instant, \textit{some but not all} components of the current estimated solution are updated. BASA allows the user to trade off memory requirements against time complexity. We develop a general methodology for proving that such algorithms converge to the fixed point of the map under study. These convergence proofs make use of weaker hypotheses than existing results. Specifically, existing convergence proofs require that the measurement noise is a zero-mean i.i.d\ sequence or a martingale difference sequence. In the present paper, we permit biased measurements, that is, measurement noises that have nonzero conditional mean. Also, all convergence results to date assume that the stochastic step sizes satisfy a probabilistic analog of the well-known Robbins-Monro conditions. We replace this assumption by a purely deterministic condition on the irreducibility of the underlying Markov processes. As specific applications to Reinforcement Learning, we analyze the temporal difference algorithm $TD(\lambda)$ for value iteration, and the $Q$-learning algorithm for finding the optimal action-value function. In both cases, we establish the convergence of these algorithms, under milder conditions than in the existing literature.
This paper studies the capacity region of asynchronous multiple access channel (MAC) with faster-thanNyquist (FTN) signaling. We first express the capacity region in the frequency domain. Next, we calculate an achievable rate region in time domain and prove that it is identical to the capacity region calculated in the frequency domain. Our analysis confirms that asynchronous transmission and FTN bring in significant gains.
Effective multi-robot teams require the ability to move to goals in complex environments in order to address real-world applications such as search and rescue. Multi-robot teams should be able to operate in a completely decentralized manner, with individual robot team members being capable of acting without explicit communication between neighbors. In this paper, we propose a novel game theoretic model that enables decentralized and communication-free navigation to a goal position. Robots each play their own distributed game by estimating the behavior of their local teammates in order to identify behaviors that move them in the direction of the goal, while also avoiding obstacles and maintaining team cohesion without collisions. We prove theoretically that generated actions approach a Nash equilibrium, which also corresponds to an optimal strategy identified for each robot. We show through extensive simulations that our approach enables decentralized and communication-free navigation by a multi-robot system to a goal position, and is able to avoid obstacles and collisions, maintain connectivity, and respond robustly to sensor noise.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191