检索结果-内蒙古大学图书馆

Asynchronous Federated Split Learning

学校读者我要写书评

暂无评论

Asynchronous Federated Split Learning

IEEE International Conference on Fog and Edge Computing (ICFEC)

作者： R. A. Albuquerque L. P. Dias Momo Ziazet K. Vandikas S. Ickin B. Jaumard C. Natalino L. Wosinska P. Monti E. Wong Computer Science and Software Engineering Concordia University Montreal Canada ER/AI Ericsson Kista Sweden GAIA Ericsson Montreal Canada Optical Networks Chalmers University of Technology Gothenburg Sweden Electrical and Electronic Engineering The University of Melbourne Melbourne Australia

ISBN: (数字)9798350361353

ISBN: (纸本)9798350361360

We propose a first Asynchronous Federated Split Learning (AFSL), to add the flexibility of asynchronous computing to the combination of federated and split learning. This amounts to designing a harmonious combination of different paradigms in order to benefit from the advantages of each of them and to reduce the impacts of their *** way, AFSL answers to the increasingly rising interest for distributed algorithms with the advent of edge computing in order to support new market segments, such as cloud gaming, immersive eXtended Reality (XR), indoor positioning, and mission critical IoT networks, with stringent requirements on latency and *** experiments are conducted on IID and non-IID datasets to investigate the added value of the asynchronous feature. Results indicate that AFSL can accelerate model learning by up to 86% without sacrificing the model’s convergence and accuracy. Indeed, not only average training times are reduced, but clients use fewer resources, a critical characteristic for devices with limited computing capabilities, e.g., in edge devices. Performance degradation can be mitigated by a careful selection of the aggregation principle. Other advantages are with AFSL training in dynamic scenarios as it provides robustness with a short recovery time by leveraging asynchronous client training.

关键词： Training Performance evaluation Accuracy Extended reality Computational modeling Mission critical systems Reliability engineering Robustness Edge computing Convergence

Random Features Model with General Convex Regularization: A Fine Grained Analysis with Precise Asymptotic Learning Curves

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Bosch, David Panahi, Ashkan Özcelikkale, Ayca Dubhashi, Devdatt Department of Data Science and AI Computer Science and Engineering Chalmers University of Technology Sweden Signals and Systems Department of Electrical Engineering Uppsala University Sweden

We compute precise asymptotic expressions for the learning curves of least squares random feature (RF) models with either a separable strongly convex regularization or the 1 regularization. We propose a novel multi-level application of the convex Gaussian min max theorem (CGMT) to overcome the traditional difficulty of finding computable expressions for random features models with correlated data. Our result takes the form of a computable 4-dimensional scalar optimization. In contrast to previous results, our approach does not require solving an often intractable proximal operator, which scales with the number of model parameters. Furthermore, we extend the universality results for the training and generalization errors for RF models to 1 regularization. In particular, we demonstrate that under mild conditions, random feature models with elastic net or 1 regularization are asymptotically equivalent to a surrogate Gaussian model with the same first and second moments. We numerically demonstrate the predictive capacity of our results, and show experimentally that the predicted test error is accurate even in the non-asymptotic *** Codes 68T07 Copyright © 2022, The Authors. All rights reserved.

关键词： Machine learning

Optimizing CDN Architectures: Multi-Metric Algorithmic Breakthroughs for Edge and Distributed Performance

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Absur, Md. Nurul Saha, Sourya Nova, Sifat Nawrin Nasif, Kazi Fahim Ahmad Nasib, Md. Rahat Ul Department of Computer Science City University of New York New York United States Department of Computer Science Chalmers University of Technology Gothenburg Sweden College of Computing and Software Engineering Kennesaw State University Georgia United States Samsung Austin Research Center AustinTX United States

A Content Delivery Network (CDN) is a powerful system of distributed caching servers that aims to accelerate content delivery, like high-definition video, IoT applications, and ultra-low-latency services, efficiently and with fast velocity. This has become of paramount importance in the post-pandemic era. Challenges arise when exponential content volume growth and scalability across different geographic locations are required. This paper investigates data-driven evaluations of CDN algorithms in dynamic server selection for latency reduction, bandwidth throttling for efficient resource management, real-time Round Trip Time analysis for adaptive routing, and programmatic network delay simulation to emulate various conditions. Key performance metrics, such as round-trip time (RTT) and CPU usage, are carefully analyzed to evaluate scalability and algorithmic efficiency through two experimental setups: a constrained edge-like local system and a scalable FABRIC testbed. The statistical validation of RTT trends, alongside CPU utilization, is presented in the results. The optimization process reveals significant trade-offs between scalability and resource consumption, providing actionable insights for effectively deploying and enhancing CDN algorithms in edge and distributed computing environments. © 2024, CC BY-NC-ND.

关键词： Scalability

Assessing the Unobserved: Enhancing Causal Inference in Sociology with Sensitivity Analysis

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Lin, Cheng Peña, Jose M. Daoud, Adel Linköping University Linköping Sweden Linköping University Linköping Sweden The Division of Data Science and Artificial Intelligence The Department of Computer Science and Engineering Chalmers University of Technology Gothenburg Sweden

Explaining social events is a primary objective of applied data-driven sociology. To achieve that objective, many sociologists use statistical causal inference to identify causality using observational studies—research context where the analyst does not control the data generating process. However, it is often challenging in observation studies to satisfy the unmeasured confounding assumption, namely, that there is no lurking third variable affecting the causal relationship of interest. By reviewing articles in American Journal of Sociology and American Sociological Review, we find that scholars tend to use a variety of strategies to avoid using causal language directly by for example refraining to disclose their causal assumptions, denoting their study as causal, and find other verbs than "causal" to denote "cause and effect." That is the case, despite the goal of their studies being explicitly explanatory. While those strategies are likely employed to navigate the publication process (getting passed the peer review phase), we argue that those strategies are likely holding back applied research. In this article, we develop a framework enabling sociologists to employ a different strategy to enhance the quality of observational studies. Our framework builds on a surprisingly simple statistical approach, sensitivity analysis: a thought-experimental framework where the analyst imagines a lever, which they can pull for probing a variety of theoretically driven statistical magnitudes of posited unmeasured confounding which in turn distorts the causal effect of interest. By pulling that lever, the analyst can identify how strong an unmeasured confounder must be to wash away the estimated causal effect. Although each sensitivity analysis method requires its own assumptions, this sort of post-hoc analysis provides underutilized tools to bound causal quantities. Extending Lundberg et al, we develop a five-step approach to how applied sociological research can incorporate sen

关键词： Sensitivity analysis

Deep Q-learning: a robust control approach

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Varga, Balázs Kulcsár, Balázs Chehreghani, Morteza Haghir Department of Electrical Engineering Chalmers University of Technology Hörsalsvägen 11 Gothenburg Sweden Department of Computer Science and Engineering Chalmers University of Technology Hörsalsvägen 11 Gothenburg Sweden

This work aims at constructing a bridge between robust control theory and reinforcement learning. Although, reinforcement learning has shown admirable results in complex control tasks, the agent's learning behaviour is opaque. Meanwhile, system theory has several tools for analyzing and controlling dynamical systems. This paper places deep Q-learning is into a control-oriented perspective to study its learning dynamics with well-established techniques from robust control. An uncertain linear time-invariant model is formulated by means of the neural tangent kernel to describe learning. This novel approach allows giving conditions for stability (convergence) of the learning and enables the analysis of the agent's behaviour in frequency-domain. The control-oriented approach makes it possible to formulate robust controllers that inject dynamical rewards as control input in the loss function to achieve better convergence properties. Three output-feedback controllers are synthesized: gain scheduling H2, dynamical H∞, and fixed-structure H∞ controllers. Compared to traditional deep Q-learning techniques, which involve several heuristics, setting up the learning agent with a control-oriented tuning methodology is more transparent and has well-established literature. The proposed approach does not use a target network and randomized replay memory. The role of the target network is overtaken by the control input, which also exploits the temporal dependency of samples (opposed to a randomized memory buffer). Numerical simulations in different OpenAI Gym environments suggest that the H∞ controlled learning can converge faster and receive higher scores (depending on the environment) compared to the benchmark Double deep Q-learning. Copyright © 2022, The Authors. All rights reserved.

关键词： Robust control

Governing the Commons: Code Ownership and Code-Clones in Large-Scale Software Development

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Sundelin, Anders Gonzalez-Huerta, Javier Torkar, Richard Wnuk, Krzysztof Department of Software Engineering Blekinge University of Technology Karlskrona371 79 Sweden Ericsson Mobile Financial Services AB Karlskrona371 23 Sweden Department of Computer Science and Engineering Chalmers University of Technology University of Gothenburg Göteborg412 96 Sweden Stellenbosch South Africa

Context: In software development organizations employing weak or collective ownership, different teams are allowed and expected to autonomously perform changes in various components. This creates diversity both in the knowledge of, and in the responsibility for, individual components. Objective: Our objective is to understand how and why different teams introduce technical debt in the form of code clones as they change different components. Method: We collected data about change size and clone introductions made by ten teams in eight components which was part of a large industrial software system. We then designed a Multi-Level Generalized Linear Model (MLGLM), to illustrate the teams’ differing behavior. Finally, we discussed the results with three development teams, plus line manager and the architect team, evaluating whether the model inferences aligned with what they expected. Responses were recorded and thematically coded. Results: The results show that teams do behave differently in different components, and the feedback from the teams indicates that this method of illustrating team behavior can be useful as a complement to traditional summary statistics of ownership. Conclusions: We find that our model-based approach produces useful visualizations of team introductions of code clones as they change different components. Practitioners stated that the visualizations gave them insights that were useful, and by comparing with an average team, inter-team comparisons can be avoided. Thus, this has the potential to be a useful feedback tool for teams in software development organizations that employ weak or collective ownership. © 2024, CC BY.

关键词：

A CONTEXTUAL COMBINATORIAL SEMI-BANDIT APPROACH TO NETWORK BOTTLENECK IDENTIFICATION

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Hoseini, Fazeleh Åkerblom, Niklas Chehreghani, Morteza Haghir Department of Computer Science and Engineering Chalmers University of Technology Gothenburg Sweden Volvo Car Corporation Gothenburg Sweden

Bottleneck identification is a challenging task in network analysis, especially when the network is not fully specified. To address this task, we develop a unified online learning framework based on combinatorial semi-bandits that performs bottleneck identification in parallel with learning the specifications of the underlying network. Within this framework, we adapt and study various combinatorial semi-bandit methods such as epsilon-greedy, LinUCB, BayesUCB, NeuralUCB, and Thompson Sampling. In addition, our framework is capable of using contextual information in the form of contextual bandits. Finally, we evaluate our framework on the real-world application of road networks and demonstrate its effectiveness in different settings. Copyright © 2022, The Authors. All rights reserved.

关键词： E-learning

Modeling PROTAC Degradation Activity with Machine Learning

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ribes, Stefano Nittinger, Eva Tyrchan, Christian Mercado, Rocío Department of Computer Science and Engineering Section for Data Science and AI Chalmers University of Technology Chalmersplatsen 4 Gothenburg412 96 Sweden BioPharmaceuticals R&D AstraZeneca Pepparedsleden 1 Mölndal431 83 Sweden

PROTACs are a promising therapeutic modality that harnesses the cell’s built-in degradation machinery to degrade specific proteins. Despite their potential, developing new PROTACs is challenging and requires significant domain expertise, time, and cost. Meanwhile, machine learning has transformed drug design and development. In this work, we present a strategy for curating open-source PROTAC data and an open-source deep learning tool for predicting the degradation activity of novel PROTAC molecules. The curated dataset incorporates important information such as pDC50 , Dmax , E3 ligase type, POI amino acid sequence, and experimental cell type. Our model architecture leverages learned embeddings from pretrained machine learning models, in particular for encoding protein sequences and cell type information. We assessed the quality of the curated data and the generalization ability of our model architecture against new PROTACs and targets via three tailored studies, which we recommend other researchers to use in evaluating their degradation activity models. In each study, three models predict protein degradation in a majority vote setting, reaching a top test accuracy of 80.8% and 0.865 ROC AUC, and a test accuracy of 62.3% and 0.604 ROC AUC when generalizing to novel protein targets. Our results are not only comparable to state-of-the-art models for protein degradation prediction, but also part of an open-source implementation which is easily reproducible and less computationally complex than existing approaches. © 2024, CC BY.

关键词：

Pure Exploration in Bandits with Linear Constraints

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Carlsson, Emil Basu, Debabrota Johansson, Fredrik D. Dubhashi, Devdatt Department of Computer Science and Engineering Chalmers University of Technology Sweden Équipe Scool Univ. Lille Inria CNRS Centrale Lille UMR 9189 CRIStAL France

We address the problem of identifying the optimal policy with a fixed confidence level in a multi-armed bandit setup, when the arms are subject to linear constraints. Unlike the standard best-arm identification problem which is well studied, the optimal policy in this case may not be deterministic and could mix between several arms. This changes the geometry of the problem which we characterize via an information-theoretic lower bound. We introduce two asymptotically optimal algorithms for this setting, one based on the Track-and-Stop method and the other based on a game-theoretic approach. Both these algorithms try to track an optimal allocation based on the lower bound and computed by a weighted projection onto the boundary of a normal cone. Finally, we provide empirical results that validate our bounds and visualize how constraints change the hardness of the problem. © 2023, CC BY.

关键词： Game theory