We consider the problem of distributed dynamic rate and channel selection in a multi-user network, in which each user selects a wireless channel and a modulation and coding scheme (corresponds to a transmission rate) ...
详细信息
We consider the problem of distributed dynamic rate and channel selection in a multi-user network, in which each user selects a wireless channel and a modulation and coding scheme (corresponds to a transmission rate) in order to maximize the network throughput. We assume that the users are cooperative, however, there is no coordination and communication among them, and the number of users in the system is unknown. We formulate this problem as a multi-player multi-armed bandit problem and propose a decentralized learning algorithm that performs almost optimal exploration of the transmission rates to learn fast. We prove that the regret of our learning algorithm with respect to the optimal allocation increases logarithmically over rounds with a leading term that is logarithmic in the number of transmission rates. Finally, we compare the performance of our learning algorithm with the state-of-the-art via simulations and show that it substantially improves the throughput and minimizes the number of collisions.
In this article, we propose several novel distributed gradient-based temporal-difference algorithms for multiagent off-policy learning of linear approximation of the value function in Markov decision processes with st...
详细信息
In this article, we propose several novel distributed gradient-based temporal-difference algorithms for multiagent off-policy learning of linear approximation of the value function in Markov decision processes with strict information structure constraints, limiting interagent communications to small neighborhoods. The algorithms are composed of the following: first, local parameter updates based on the single-agent off-policy gradient temporal-difference learning algorithms, including the eligibility traces with state-dependent parameters and, second, linear stochastic time-varying consensus schemes, represented by directed graphs. The proposed algorithms differ in their form, definition of eligibility traces, selection of time scales, and the way of incorporating consensus iterations. The main contribution of this article is a convergence analysis based on the general properties of the underlying Feller-Markov processes and the stochastic time-varying consensus model. We prove under general assumptions that the parameter estimates generated by all the proposed algorithms weakly converge to the corresponding ordinary differential equations with precisely defined invariant sets. It is demonstrated how the adopted methodology can be applied to temporal-difference algorithms under weaker information structure constraints. The variance reduction effect of the proposed algorithms is demonstrated by formulating and analyzing an asymptotic stochastic differential equation. Specific guidelines for the communication network design are provided. The algorithms' superior properties are illustrated by characteristic simulation results.
Stochastic gradient Langevin dynamics (SGLD) and stochastic gradient Hamiltonian Monte Carlo (SGHMC) are two popular Markov Chain Monte Carlo (MCMC) algorithms for Bayesian inference that can scale to large datasets, ...
详细信息
Stochastic gradient Langevin dynamics (SGLD) and stochastic gradient Hamiltonian Monte Carlo (SGHMC) are two popular Markov Chain Monte Carlo (MCMC) algorithms for Bayesian inference that can scale to large datasets, allowing to sample from the posterior distribution of the parameters of a statistical model given the input data and the prior distribution over the model parameters. However, these algorithms do not apply to the decentralized learning setting, when a network of agents are working collaboratively to learn the parameters of a statistical model without sharing their individual data due to privacy reasons or communication constraints. We study two algorithms: decentralized SGLD (DE-SGLD) and decentralized SGHMC (DE-SGHMC) which are adaptations of SGLD and SGHMC methods that allow scaleable Bayesian inference in the decentralized setting for large datasets. We show that when the posterior distribution is strongly log-concave and smooth, the iterates of these algorithms converge linearly to a neighborhood of the target distribution in the 2-Wasserstein distance if their parameters are selected appropriately. We illustrate the efficiency of our algorithms on decentralized Bayesian linear regression and Bayesian logistic regression problems.
Phylogenetic analyses can also enable insight into evolutionary and ecological dynamics such as selection pressure and frequency dependent selection in digital evolution systems. Traditionally, digital evolution syste...
详细信息
ISBN:
(纸本)9781450392686
Phylogenetic analyses can also enable insight into evolutionary and ecological dynamics such as selection pressure and frequency dependent selection in digital evolution systems. Traditionally, digital evolution systems have recorded data for phylogenetic analyses through perfect tracking where each birth event is recorded in a centralized data structures. This approach, however, does not easily scale to distributed computing environments where evolutionary individuals may migrate between a large number of disjoint processing elements. To provide for phylogenetic analyses in these environments, we propose an approach to infer phylogenies via heritable genetic annotations rather than directly track them. We introduce a "hereditary stratigraphy" algorithm that enables efficient, accurate phylogenetic reconstruction with tunable, explicit trade-offs between annotation memory footprint and reconstruction accuracy. This approach can estimate, for example, MRCA generation of two genomes within 10% relative error with 95% confidence up to a depth of a trillion generations with genome annotations smaller than a kilobyte. We also simulate inference over known lineages, recovering up to 85.70% of the information contained in the original tree using a 64-bit annotation.
We consider the computation offloading problem in an edge computing system in which an operator jointly manages wireless and computing resources across devices that make their offloading decisions autonomously with th...
详细信息
We consider the computation offloading problem in an edge computing system in which an operator jointly manages wireless and computing resources across devices that make their offloading decisions autonomously with the objective to minimize their own completion times. We develop a game theoretical model of the interaction between the devices and an operator that can implement one of two resource allocation policies, a cost minimizing or a time fair resource allocation policy. We express the optimal cost minimizing resource allocation policy in closed form and prove the existence of Stackelberg equilibria for both resource allocation policies. We propose two efficient decentralized algorithms that devices can use for computing equilibria of offloading decisions under the cost minimizing and the time fair resource allocation policies. We establish bounds on the price of anarchy of the games played by the devices and by doing so we show that the proposed algorithms have bounded approximation ratios. Our simulation results show that the cost minimizing resource allocation policy can achieve significantly lower completion times than the time fair allocation policy. At the same time, the convergence time of the proposed algorithms is approximately linear in the number of devices, and thus they could be effectively implemented for edge computing resource management.
In this letter, we study decentralized stochastic optimization to minimize a sum of smooth and strongly convex cost functions when the functions are distributed over a directed network of nodes. In contrast to the exi...
详细信息
In this letter, we study decentralized stochastic optimization to minimize a sum of smooth and strongly convex cost functions when the functions are distributed over a directed network of nodes. In contrast to the existing work, we use gradient tracking to improve certain aspects of the resulting algorithm. In particular, we propose the S-ADDOPT algorithm that assumes a stochastic first-order oracle at each node and show that for a constant step-size alpha, each node converges linearly inside an error ball around the optimal solution, the size of which is controlled by alpha. For decaying step-sizes O(1/k), we show that S-ADDOPT reaches the exact solution sublinearly at O(1/k) and its convergence is asymptotically network-independent. Thus the asymptotic behavior of S-ADDOPT is comparable to the centralized stochastic gradient descent. Numerical experiments over both strongly convex and non-convex problems illustrate the convergence behavior and the performance comparison of the proposed algorithm.
Communication connectivity is desirable for the safe and efficient operation of multi-robot systems. While decentralized algorithms for connectivity mainte-nance have been explored in recent literature, the majority o...
详细信息
Communication connectivity is desirable for the safe and efficient operation of multi-robot systems. While decentralized algorithms for connectivity mainte-nance have been explored in recent literature, the majority of these works do not account for robot motion and sensing uncertainties. These uncertainties are inherent in practical robots and result in robots deviating from their desired positions which could potentially result in a loss of connectivity. In this paper, we present a decentralized connectivity maintenance algorithm accounting for robot motion and sensing uncertainties (DCMU). We, first, propose a novel weighted graph definition for the multi-robot system that accounts for the afore-mentioned uncertainties along with realistic connectivity constraints such as line-of-sight connectivity and collision avoidance. We, then, design a decen-tralized gradient-based controller for connectivity maintenance with which we derive the gradients of the weighted graph edge weights required for computing the control. Finally, we perform multiple simulations to validate the connectiv-ity maintenance performance of our DCMU algorithm under robot motion and sensing uncertainties, showing an improvement compared to previous work.
In this paper, we propose a decentralized first-order stochastic optimization method Push-SAGA for finite-sum minimization over a strongly connected directed graph. This method features local variance reduction to rem...
详细信息
ISBN:
(纸本)9781728176055
In this paper, we propose a decentralized first-order stochastic optimization method Push-SAGA for finite-sum minimization over a strongly connected directed graph. This method features local variance reduction to remove the uncertainty caused by random sampling of the local gradients, global gradient tracking to address the distributed nature of the data, and push-sum consensus to handle the imbalance caused by the directed nature of the underlying graph. We show that, for a sufficiently small step-size, Push-SAGA linearly converges to the optimal solution for smooth and strongly convex problems, making it the first linearly-convergent stochastic algorithm over arbitrary strongly-connected directed graphs. We illustrate the behavior and convergence properties of Push-SAGA with the help of numerical experiments for strongly convex and non-convex problems.
We develop new statistics for robustly filtering corrupted keypoint matches in the structure from motion pipeline. The statistics are based on consistency constraints that arise within the clustered structure of the g...
详细信息
ISBN:
(纸本)9781665426886
We develop new statistics for robustly filtering corrupted keypoint matches in the structure from motion pipeline. The statistics are based on consistency constraints that arise within the clustered structure of the graph of keypoint matches. The statistics are designed to give smaller values to corrupted matches and than uncorrupted matches. These new statistics are combined with an iterative reweighting scheme to filter keypoints, which can then be fed into any standard structure from motion pipeline. This filtering method can be efficiently implemented and scaled to massive datasets as it only requires sparse matrix multiplication. We demonstrate the efficacy of this method on synthetic and real structure from motion datasets and show that it achieves state-of-the-art accuracy and speed in these tasks.
Multi-electrode arrays such as "Neuropixels" probes enable the study of neuronal voltage signals at high temporal and single-cell spatial resolution. However, in vivo recordings from these devices often expe...
详细信息
ISBN:
(纸本)9781728176055
Multi-electrode arrays such as "Neuropixels" probes enable the study of neuronal voltage signals at high temporal and single-cell spatial resolution. However, in vivo recordings from these devices often experience some shifting of the probe (due e.g. to animal movement), resulting in poorly localized voltage readings that in turn can corrupt estimates of neural activity. We introduce a new registration method to partially correct for this motion. In contrast to previous template-based registration methods, the proposed approach is decentralized, estimating shifts of the data recorded in multiple timebins with respect to one another, and then extracting a global registration estimate from the resulting estimated shift matrix. We find that the resulting decentralized registration is more robust and accurate than previous templatebased approaches applied to both simulated and real data, but nonetheless some significant non-stationarity in the recovered neural activity remains that should be accounted for by downstream processing pipelines. Open source code is available at https://***/evarol/NeuropixelsRegistration.
暂无评论