ACM Transactions on Autonomous and Adaptive Systems最新文献

Smart homes are Cyber-Physical Systems (CPS) where multiple devices and controllers cooperate to achieve high-level goals. Causal knowledge on relations between system entities is essential for enabling system self-adaption to dynamic changes. As house configurations are diverse, this knowledge is difficult to obtain. In previous work, we proposed to generate Causal Bayesian Networks (CBN) as follows. Starting with considering all possible relations, we progressively discarded non-correlated variables. Next, we identified causal relations from the remaining correlations by employing “do-operations”. The obtained CBN could then be employed for causal inference. The main challenges of this approach included: “non-doable variables” and limited scalability. To address these issues, we propose three extensions: i) early pruning weakly correlated relations to reduce the number of required do-operations; ii) introducing aggregate variables that summarize relations between weakly-coupled sub-systems; iii) applying the method a second time to perform indirect do interventions and handle non-doable relations. We illustrate and evaluate the efficiency of these contributions via examples from the smart home and power grid domain. Our proposal leads to a decrease in the number of operations required to learn the CBN and in an increased accuracy of the learned CBN, paving the way towards applications in large CPS.

The concept of pulse-coupled oscillators for self-organized synchronization has been applied to wireless systems. Putting theory into practice, however, faces certain obstacles, particularly in radio technologies that cannot implement pulses but use common messages for interactions between nodes. This raises the question of how to deal with interference between messages. We show that interference can disturb the synchronization process and propose low-complex, randomization-based techniques to address this issue. First, we demonstrate that randomly switching between two transmit power levels (without increasing the average power) can expedite synchronization. The high-power transmissions temporarily boost network connectivity with negligible impact on the average interference. Second, we reduce interference by blindly distributing the messages over the entire oscillator cycle. Instead of using a fixed oscillator phase at which the pulses are sent, each node chooses its own, randomly selected phase to send a synchronization message. This node-specific “fire phase” is contained in the message to permit others to compute the timing. Third, we suggest that such interference management can also be beneficial for other synchronization techniques and validate this claim using Glossy as an example. Our insights may contribute to feasible solutions for self-organized wireless synchronization. Further work is needed to gain a comprehensive understanding of the effects of randomization and to develop algorithms for the adaptability of local parameters.

With the advent of 5G+ services, it has become increasingly convenient for mobile users to enjoy high quality multimedia content from CDN driven streaming and catch-up TV services (Netflix, iPlayer) in the (post-)COVID over-the-top (OTT) content rush. To relieve ISP owned fixed-line networks from CDN streamed multimedia traffic, system ideas (e.g., Wi-Stitch in [45]) have been proposed to (a) leverage 5G services and enable consumers to share cached multimedia content at the edge, and (b) consequently, and more importantly, reduce IP traffic at the core network. Unfortunately, given that contemporary multimedia content might be a monetised asset, these ideas do not take this important fact into account for shared content.We present EdgeMart - a content provider federated, and computationally sustainable networked (graphical) market economy for paid-sharing of cached licensed (OTT) content with autonomous users of a wireless edge network (WEN). EdgeMart is a unique oligopoly multimedia market (economy) that comprises competing networked sub-markets of non-cooperative content sellers/buyers - each sub-market consisting of a single buyer connected (networked) to only a subset of sellers. We prove that for any WEN-supported supply-demand topology, a pure strategy EdgeMart equilibrium exists that is (a) nearly efficient (in a microeconomic sense) indicating economy sustainability, (b) robust to edge user entry/exit, and (c) can be reached in poly-time (indicating computational sustainability). In addition, we experimentally show that for physical WENs of varying densities, a rationally selfish EdgeMart economy induces similar orders of multimedia IP traffic savings when compared to the ideal (relatively less practical), altruistic, and non-monetized “economy” implemented atop the recently introduced Wi-Stitch WEN-based content trading architecture. Moreover, the EdgeMart concept helps envision a regulated edge economy of opportunistic (pay per licensed file) client services for commercial OTT platforms.

Given the popular presupposition of human reasoning as the standard for learning and decision making, there have been significant efforts and a growing trend in research to replicate these innate human abilities in artificial systems. As such, topics including Game Theory, Theory of Mind, and Machine Learning, among others, integrate concepts that are assumed components of human reasoning. These serve as techniques to replicate and understand the behaviors of humans. In addition, next-generation autonomous and adaptive systems will largely include AI agents and humans working together as teams. To make this possible, autonomous agents will require the ability to embed practical models of human behavior, allowing them not only to replicate human models as a technique to “learn” but also to understand the actions of users and anticipate their behavior, so as to truly operate in symbiosis with them. The main objective of this article is to provide a succinct yet systematic review of important approaches in two areas dealing with quantitative models of human behaviors. Specifically, we focus on (i) techniques that learn a model or policy of behavior through exploration and feedback, such as Reinforcement Learning, and (ii) directly model mechanisms of human reasoning, such as beliefs and bias, without necessarily learning via trial and error.

In the implementation of deep reinforcement learning (DRL), action persistence strategies are often adopted so agents maintain their actions for a fixed or variable number of steps. The choice of the persistent duration for agent actions usually has notable effects on the performance of reinforcement learning algorithms. Aiming at the research gap of global dynamic optimal action persistence and its application in multi-agent systems, we propose a novel framework: global dynamic action persistence (GLDAP), which achieves global action persistence adaptation for deep reinforcement learning. We introduce a closed-loop method that is used to learn the estimated value and the corresponding policy of each candidate action persistence. Our experiment shows that GLDAP achieves an average of 2.5%~90.7% performance improvement and 3~20 times higher sampling efficiency over several baselines across various single-agent and multi-agent domains. We also validate the ability of GLDAP to determine the optimal action persistence through multiple experiments.

With the rise of new technologies, such as Edge computing, Internet of Things, Smart Cities, and Smart Grids, there is a growing need for multi-agent systems (MAS) approaches. Designing multi-agent systems is challenging, and doing this in an automated way is even more so. To address this, we propose a new framework, Evolved Gossip Contracts (EGC). It builds on Gossip Contracts (GC), a decentralised cooperation protocol that is used as the communication mechanism to facilitate self-organisation in a cooperative MAS. GC has several methods that are implemented uniquely, depending on the goal the MAS aims to achieve. The EGC framework uses evolutionary computing to search for the best implementation of these methods. To evaluate EGC, it was used to solve a classical NP-hard optimisation problem, the Bin Packing Problem (BPP). The experimental results show that EGC successfully discovered a decentralised strategy to solve the BPP, which is better than two classical heuristics on test cases similar to those on which it was trained; the improvement is statistically significant. EGC is the first framework that leverages evolutionary computing to semi-automate the discovery of a communication protocol for a MAS that has been shown to be effective at solving an NP-hard problem.