An official website of the United States government
The advances in deep neural networks (DNN) have significantly enhanced real-time detection of anomalous data in IoT applications. However, the complexity-accuracy-delay dilemma persists: Complex DNN models offer higher accuracy, but typical IoT devices can barely afford the computation load, and the remedy of offloading the load to the cloud incurs long delay. In this article, we address this challenge by proposing an adaptive anomaly detection scheme with hierarchical edge computing (HEC). Specifically, we first construct multiple anomaly detection DNN models with increasing complexity and associate each of them to a corresponding HEC layer. Then, we design an adaptive model selection scheme that is formulated as a contextual-bandit problem and solved by using a reinforcement learning policy network . We also incorporate a parallelism policy training method to accelerate the training process by taking advantage of distributed models. We build an HEC testbed using real IoT devices and implement and evaluate our contextual-bandit approach with both univariate and multivariate IoT datasets. In comparison with both baseline and state-of-the-art schemes, our adaptive approach strikes the best accuracy-delay tradeoff on the univariate dataset and achieves the best accuracy and F1-score on the multivariate dataset with only negligibly longer delay than the best (but inflexible) scheme.
more »
« less
Persistent and responsive collective motion with adaptive time delay
It is beneficial for collective structures to simultaneously have high persistence to environmental noise and high responsivity to nontrivial external stimuli. However, without the ability to differentiate useful information from noise, there is always a tradeoff between persistence and responsivity within the collective structures. To address this, we propose adaptive time delay inspired by the adaptive behavior observed in the school of fish. This strategy is tested using particles powered by optothermal fields coupled with an optical feedback-control system. By applying the adaptive time delay with a proper threshold, we experimentally observe the responsivity of the collective structures enhanced by approximately 1.6 times without sacrificing persistence. Furthermore, we integrate adaptive time delay with long-distance transportation and obstacle-avoidance capabilities to prototype adaptive swarm microrobots. This research demonstrates the potential of adaptive time delay to address the persistence-responsivity tradeoff and lays the foundation for intelligent swarm micro/nanorobots operating in complex environments.
more »
« less
- Award ID(s):
- 2001650
- PAR ID:
- 10563129
- Publisher / Repository:
- American Association for the Advancement of Science
- Date Published:
- Journal Name:
- Science Advances
- Volume:
- 10
- Issue:
- 14
- ISSN:
- 2375-2548
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Honey bees (Apis mellifera L.) localize the queen and aggregate into a swarm by forming a collective scenting network to directionally propagate volatile pheromone signals. Previous experiments show the robustness of this communication strategy in the presence of physical obstacles that partially block pheromone flow and the path to the queen. Specifically, there is a delay in the formation of the scenting network and aggregation compared to a simple environment without perturbations. To better understand the effect of obstacles beyond temporal dynamics, we use the experimental results as inspiration to explore how the behavioral parameter space of collective scenting responds to obstacle. We extend an agent-based model previously developed for a simple environment to account for the presence of physical obstacles. We study how individual agents with simple behavioral rules for scenting and following concentration gradients can give rise to collective localization and swarming. We show that the bees are capable of navigating the more complex environment with a physical obstacle to localize the queen and aggregate around her, but their range of behavioral parameters is more limited and less flexible as a result of the spatial density heterogeneity in the bees imposed by the obstacle.more » « less
-
There are a number of forums where people participate under pseudonyms. One example is peer review, where the identity of reviewers for any paper is confidential. When participating in these forums, people frequently engage in batching: executing multiple related tasks (e.g., commenting on multiple papers) at nearly the same time. Our empirical analysis shows that batching is common in two applications we consider -- peer review and Wikipedia edits. In this paper, we identify and address the risk of deanonymization arising from linking batched tasks. To protect against linkage attacks, we take the approach of adding delay to the posting time of batched tasks. We first show that under some natural assumptions, no delay mechanism can provide a meaningful differential privacy guarantee. We therefore propose a one-sided formulation of differential privacy for protecting against linkage attacks. We design a mechanism that adds zero-inflated uniform delay to events and show it can preserve privacy. We prove that this noise distribution is in fact optimal in minimizing expected delay among mechanisms adding independent noise to each event, thereby establishing the Pareto frontier of the trade-off between the expected delay for batched and unbatched events. Finally, we conduct a series of experiments on Wikipedia and Bitcoin data that corroborate the practical utility of our algorithm in obfuscating batching without introducing onerous delay to a system.more » « less
-
In this paper, a distributed swarm control problem is studied for large-scale multi-agent systems (LS-MASs). Different than classical multi-agent systems, an LS-MAS brings new challenges to control design due to its large number of agents. It might be more difficult for developing the appropriate control to achieve complicated missions such as collective swarming. To address these challenges, a novel mixed game theory is developed with a hierarchical learning algorithm. In the mixed game, the LS-MAS is represented as a multi-group, large-scale leader–follower system. Then, a cooperative game is used to formulate the distributed swarm control for multi-group leaders, and a Stackelberg game is utilized to couple the leaders and their large-scale followers effectively. Using the interaction between leaders and followers, the mean field game is used to continue the collective swarm behavior from leaders to followers smoothly without raising the computational complexity or communication traffic. Moreover, a hierarchical learning algorithm is designed to learn the intelligent optimal distributed swarm control for multi-group leader–follower systems. Specifically, a multi-agent actor–critic algorithm is developed for obtaining the distributed optimal swarm control for multi-group leaders first. Furthermore, an actor–critic–mass method is designed to find the decentralized swarm control for large-scale followers. Eventually, a series of numerical simulations and a Lyapunov stability proof of the closed-loop system are conducted to demonstrate the performance of the developed scheme.more » « less
-
Abstract Development of microbial communities is a complex multiscale phenomenon with wide-ranging biomedical and ecological implications. How biological and physical processes determine emergent spatial structures in microbial communities remains poorly understood due to a lack of simultaneous measurements of gene expression and cellular behaviour in space and time. Here we combined live-cell microscopy with a robotic arm for spatiotemporal sampling, which enabled us to simultaneously acquire phenotypic imaging data and spatiotemporal transcriptomes duringBacillus subtilisswarm development. Quantitative characterization of the spatiotemporal gene expression patterns revealed correlations with cellular and collective properties, and phenotypic subpopulations. By integrating these data with spatiotemporal metabolome measurements, we discovered a spatiotemporal cross-feeding mechanism fuelling swarm development: during their migration, earlier generations deposit metabolites which are consumed by later generations that swarm across the same location. These results highlight the importance of spatiotemporal effects during the emergence of phenotypic subpopulations and their interactions in bacterial communities.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1126/sciadv.adk3914
