As a promising method for central model training on decentralized device data without compromising user privacy, federated learning (FL) is becoming more and more popular in Internet-of-Things (IoT) design. However, due to limited computing and memory resources of devices that restrict the capabilities of hosted deep learning models, existing FL approaches for artificial intelligence IoT (AIoT) applications suffer from inaccurate prediction results. To address this problem, this article presents a collaborativeBig.Littlebranch architecture to enable efficient FL for AIoT applications. Inspired by the architecture of BranchyNet which has multiple prediction branches, our approach deploys deep neural network (DNN) models across both cloud and AIoT devices. OurBig.Littlebranch model has two branches, where the big branch is deployed on cloud for strengthened prediction accuracy, and the little branches are used to fit for AIoT devices. When AIoT devices cannot make the prediction with high confidence using local little branches, they will resort to the big branch for further inference. To increase both prediction accuracy and early exit rate ofBig.Littlebranch model, we propose a two-stage training and coinference scheme, which considers the local characteristics of AIoT scenarios. Comprehensive experiment results obtained from a real AIoT environment demonstrate the efficiency and effectiveness of our approach in terms of prediction accuracy and average inference time.

Aiming at high performance, more and more Cyber-Physical Systems (CPSs) adopt Multiprocessor System-on-Chips (MPSoCs) as computation units. However, due to increasing integration of transistors on a die, the power densities together with performance variations of MPSoC chips have been increasing dramatically. Consequently, the MPSoC-based CPSs might become unsustainable and unreliable. Although various Task Allocation and Scheduling (TAS) heuristics have been proposed to minimize the hotspot time (i.e., duration of thermal emergency) and energy consumption of MPSoC designs, few of them can guarantee the highest performance yield under process variations without violating energy, thermal and timing constraints. To address these challenges, this paper proposes a novel energy- and thermal-aware TAS evaluation and optimization framework. Based on statistical model checking techniques, our approach enables accurate modeling and reasoning of the performance yield of real-time MPSoC designs under joint energy and thermal constraints. To enable system-level design space exploration, we propose a regression analysis-based method that can drastically reduce the overall exploration efforts. Experimental results show that our fully-automated approach can not only allow accurate sustainability-oriented reasoning of TAS solutions under specified thermal and energy constraints, but also enable the quick search of optimal TAS solutions on different MPSoC architectures with the highest performance yield.

Branch-and-bound approaches are promising in pruning fruitless search space during the resource constrained scheduling. However, such approaches only compare the estimated upper and lower bounds of an incomplete schedule to the length of the best feasible schedule at that iteration, which does not fully exploit the potential of the pruning during the search. Aiming to improve the performance of resource constrained scheduling, this paper proposes a parallel structure-aware pruning approach that can traverse the search space significantly faster than state-of-the-art branch-and-bound techniques. This paper makes three major contributions: i) it proposes an efficient pruning technique using the structural scheduling information of the obtained best feasible schedules; ii) it investigates how to perform parallel search to enable efficient multi-directional search and generation of effective fences by tuning the operation enumeration order; and iii) it presents a framework that supports the sharing of minimum upper-bound and fence information among different search tasks to enable efficient parallel structure-aware pruning. The experimental results demonstrate that our parallel pruning approach can drastically reduce the overall resource constrained scheduling time under a wide variety of resource constraints.