Recent trends like the Internet of Things (IoT) suggest a vision of dense and multi-scale deployments of computing devices in nearly all kinds of environments. A prominent engineering challenge revolves around programming the collective adaptive behaviour of such computational ecosystems. This requires abstractions able to capture concepts like ensembles (dynamic groups of cooperating devices) and collective tasks (joint activities carried out by ensembles). In this work, we consider collections of devices interacting with neighbours and that execute in nearly-synchronised sense-compute-interact rounds, where the computation is given by a single program mapping sensing values and incoming messages to output and outcoming messages. To support programming whole computational collectives, we propose the abstraction of a distributed collective process, which can be used to define at once the ensemble formation logic and its collective task. We formalise the abstraction in the eXchange Calculus (XC), a core functional language based on neighbouring values (maps from neighbours to values) where state and interaction is handled through a single primitive, exchange, and provide a corresponding implementation in the FCPP language. Then, we exercise distributed collective processes using two case studies: multi-hop message propagation and distributed monitoring of spatial properties. Finally, we discuss the features of the abstraction and its suitability for different kinds of distributed computing applications.
相關內容
To enable large-scale and efficient deployment of artificial intelligence (AI), the combination of AI and edge computing has spawned Edge Intelligence, which leverages the computing and communication capabilities of end devices and edge servers to process data closer to where it is generated. A key technology for edge intelligence is the privacy-protecting machine learning paradigm known as Federated Learning (FL), which enables data owners to train models without having to transfer raw data to third-party servers. However, FL networks are expected to involve thousands of heterogeneous distributed devices. As a result, communication efficiency remains a key bottleneck. To reduce node failures and device exits, a Hierarchical Federated Learning (HFL) framework is proposed, where a designated cluster leader supports the data owner through intermediate model aggregation. Therefore, based on the improvement of edge server resource utilization, this paper can effectively make up for the limitation of cache capacity. In order to mitigate the impact of soft clicks on the quality of user experience (QoE), the authors model the user QoE as a comprehensive system cost. To solve the formulaic problem, the authors propose a decentralized caching algorithm with federated deep reinforcement learning (DRL) and federated learning (FL), where multiple agents learn and make decisions independently
The stochastic block model (SBM) is a generalization of the Erd\H{o}s--R\'enyi model of random graphs that describes the interaction of a finite number of distinct communities. In sparse Erd\H{o}s--R\'enyi graphs, it is known that a linear-time algorithm of Karp and Sipser achieves near-optimal matching sizes asymptotically almost surely, giving a law-of-large numbers for the matching sizes of such graphs in terms of solutions to an ODE. We provide an extension of this analysis, identifying broad ranges of stochastic block model parameters for which the Karp--Sipser algorithm achieves near-optimal matching sizes, but demonstrating that it cannot perform optimally on general SBM instances. We also consider the problem of constructing a matching online, in which the vertices of one half of a bipartite stochastic block model arrive one-at-a-time, and must be matched as they arrive. We show that the competitive ratio lower bound of 0.837 found by Mastin and Jaillet for the Erd\H{o}s--R\'enyi case is tight whenever the expected degrees in all communities are equal. We propose several linear-time algorithms for online matching in the general stochastic block model, but prove that despite very good experimental performance, none of these achieve online asymptotic optimality.
Thanks to technologies such as virtual network function the Fifth Generation (5G) of mobile networks dynamically allocate resources to different types of users in an on-demand fashion. Virtualization extends up to the 5G core, where software-defined networks and network slicing implement a customizable environment. These technologies can be controlled via application programming interfaces and web technologies, inheriting hence their security risks and settings. An attacker exploiting vulnerable implementations of the 5G core may gain privileged control of the network assets and disrupt its availability. However, there is currently no security assessment of the web security of the 5G core network. In this paper, we present the first security assessment of the 5G core from a web security perspective. We use the STRIDE threat modeling approach to define a complete list of possible threat vectors and associated attacks. Thanks to a suite of security testing tools, we cover all of these threats and test the security of the 5G core. In particular, we test the three most relevant open-source 5G core implementations, i.e., Open5GS, Free5Gc, and OpenAirInterface. Our analysis shows that all these cores are vulnerable to at least two of our identified attack vectors, demanding increased security measures in the development of future 5G core networks.
The computing in the network (COIN) paradigm is a promising solution that leverages unused network resources to perform tasks to meet computation-demanding applications, such as the metaverse. In this vein, we consider the partial computation offloading problem in the metaverse for multiple subtasks in a COIN environment to minimize energy consumption and delay while dynamically adjusting the offloading policy based on the changing computational resource status. The problem is NP-hard, and we transform it into two subproblems: the task-splitting problem (TSP) on the user side and the task-offloading problem (TOP) on the COIN side. We model the TSP as an ordinal potential game and propose a decentralized algorithm to obtain its Nash equilibrium (NE). Then, we model the TOP as a Markov decision process and propose the double deep Q-network (DDQN) to solve for the optimal offloading policy. Unlike the conventional DDQN algorithm, where intelligent agents sample offloading decisions randomly within a certain probability, the COIN agent explores the NE of the TSP and the deep neural network. Finally, the simulation results reveal that the proposed model approach allows the COIN agent to update its policies and make more informed decisions, leading to improved performance over time compared to the traditional baseline
The Multilevel Fast Multipole Algorithm (MLFMA) has known applications in scientific modeling in the fields of telecommunications, physics, mechanics, and chemistry. Accelerating calculation of far-field using GPUs and GPU clusters for large-scale problems has been studied for more than a decade. The acceleration of the Near Field Computation (P2P operator) however was less of a concern because it does not face the challenges of distributed processing which does far field. This article proposes a modification of the P2P algorithm and uses performance models to determine its optimality criteria. By modeling the speedup, we found that making threads independence by creating redundancy in the data makes the algorithm for lower dense (higher frequency) problems nearly 13 times faster than non-redundant mode.
Contact-rich manipulation tasks often exhibit a large sim-to-real gap. For instance, industrial assembly tasks frequently involve tight insertions where the clearance is less than 0.1 mm and can even be negative when dealing with a deformable receptacle. This narrow clearance leads to complex contact dynamics that are difficult to model accurately in simulation, making it challenging to transfer simulation-learned policies to real-world robots. In this paper, we propose a novel framework for robustly learning manipulation skills for real-world tasks using simulated data only. Our framework consists of two main components: the "Force Planner" and the "Gain Tuner". The Force Planner plans both the robot motion and desired contact force, while the Gain Tuner dynamically adjusts the compliance control gains to track the desired contact force during task execution. The key insight is that by dynamically adjusting the robot's compliance control gains during task execution, we can modulate contact force in the new environment, thereby generating trajectories similar to those trained in simulation and narrowing the sim-to-real gap. Experimental results show that our method, trained in simulation on a generic square peg-and-hole task, can generalize to a variety of real-world insertion tasks involving narrow and negative clearances, all without requiring any fine-tuning. Videos are available at //dynamic-compliance.github.io.
The dominating NLP paradigm of training a strong neural predictor to perform one task on a specific dataset has led to state-of-the-art performance in a variety of applications (eg. sentiment classification, span-prediction based question answering or machine translation). However, it builds upon the assumption that the data distribution is stationary, ie. that the data is sampled from a fixed distribution both at training and test time. This way of training is inconsistent with how we as humans are able to learn from and operate within a constantly changing stream of information. Moreover, it is ill-adapted to real-world use cases where the data distribution is expected to shift over the course of a model's lifetime. The first goal of this thesis is to characterize the different forms this shift can take in the context of natural language processing, and propose benchmarks and evaluation metrics to measure its effect on current deep learning architectures. We then proceed to take steps to mitigate the effect of distributional shift on NLP models. To this end, we develop methods based on parametric reformulations of the distributionally robust optimization framework. Empirically, we demonstrate that these approaches yield more robust models as demonstrated on a selection of realistic problems. In the third and final part of this thesis, we explore ways of efficiently adapting existing models to new domains or tasks. Our contribution to this topic takes inspiration from information geometry to derive a new gradient update rule which alleviate catastrophic forgetting issues during adaptation.
Vast amount of data generated from networks of sensors, wearables, and the Internet of Things (IoT) devices underscores the need for advanced modeling techniques that leverage the spatio-temporal structure of decentralized data due to the need for edge computation and licensing (data access) issues. While federated learning (FL) has emerged as a framework for model training without requiring direct data sharing and exchange, effectively modeling the complex spatio-temporal dependencies to improve forecasting capabilities still remains an open problem. On the other hand, state-of-the-art spatio-temporal forecasting models assume unfettered access to the data, neglecting constraints on data sharing. To bridge this gap, we propose a federated spatio-temporal model -- Cross-Node Federated Graph Neural Network (CNFGNN) -- which explicitly encodes the underlying graph structure using graph neural network (GNN)-based architecture under the constraint of cross-node federated learning, which requires that data in a network of nodes is generated locally on each node and remains decentralized. CNFGNN operates by disentangling the temporal dynamics modeling on devices and spatial dynamics on the server, utilizing alternating optimization to reduce the communication cost, facilitating computations on the edge devices. Experiments on the traffic flow forecasting task show that CNFGNN achieves the best forecasting performance in both transductive and inductive learning settings with no extra computation cost on edge devices, while incurring modest communication cost.
Deep neural networks (DNNs) are successful in many computer vision tasks. However, the most accurate DNNs require millions of parameters and operations, making them energy, computation and memory intensive. This impedes the deployment of large DNNs in low-power devices with limited compute resources. Recent research improves DNN models by reducing the memory requirement, energy consumption, and number of operations without significantly decreasing the accuracy. This paper surveys the progress of low-power deep learning and computer vision, specifically in regards to inference, and discusses the methods for compacting and accelerating DNN models. The techniques can be divided into four major categories: (1) parameter quantization and pruning, (2) compressed convolutional filters and matrix factorization, (3) network architecture search, and (4) knowledge distillation. We analyze the accuracy, advantages, disadvantages, and potential solutions to the problems with the techniques in each category. We also discuss new evaluation metrics as a guideline for future research.
Deep neural networks (DNNs) have been found to be vulnerable to adversarial examples resulting from adding small-magnitude perturbations to inputs. Such adversarial examples can mislead DNNs to produce adversary-selected results. Different attack strategies have been proposed to generate adversarial examples, but how to produce them with high perceptual quality and more efficiently requires more research efforts. In this paper, we propose AdvGAN to generate adversarial examples with generative adversarial networks (GANs), which can learn and approximate the distribution of original instances. For AdvGAN, once the generator is trained, it can generate adversarial perturbations efficiently for any instance, so as to potentially accelerate adversarial training as defenses. We apply AdvGAN in both semi-whitebox and black-box attack settings. In semi-whitebox attacks, there is no need to access the original target model after the generator is trained, in contrast to traditional white-box attacks. In black-box attacks, we dynamically train a distilled model for the black-box model and optimize the generator accordingly. Adversarial examples generated by AdvGAN on different target models have high attack success rate under state-of-the-art defenses compared to other attacks. Our attack has placed the first with 92.76% accuracy on a public MNIST black-box attack challenge.
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