This book ?nds new things to do with an old idea. The proofs-as-programs paradigm constitutes a set of approaches to developing programs from proofs in constructive logic. It has been over thirty years since the parad...
详细信息
ISBN:
(数字)9780387281834
ISBN:
(纸本)9780387237596;9781441920140
This book ?nds new things to do with an old idea. The proofs-as-programs paradigm constitutes a set of approaches to developing programs from proofs in constructive logic. It has been over thirty years since the paradigm was ?rst conceived. At that time, there was a belief that proofs-as-programs had the - tential for practical application to semi-automated software development. I- tial applications were mostly concerned with ?ne-grain, mathematical program synthesis. For various reasons, research interest in the area eventually tended toward more theoretic issues of constructive logic and type theory. However, in recent years, the situation has become more balanced, and there is increasingly active research in applying constructive techniques to industrial-scale, complex softwareengineering problems. Thismonographdetailsseveralimportantadvancesinthisdirectionofpr- tical proofs-as-programs. One of the central themes of the book is a general, abstract framework for developing new systems of program synthesis by adapting proofs-as-programs to new contexts. Framework-oriented approaches that facilitate analogous - proaches to building systems for solving particular problems have been popular and successful. Thesemethodsarehelpful asthey providea formal toolbox that enablesa“roll-your-own”*** framework will have a similar impact. The framework is demonstrated by example. We will give two novel - plications of proofs-as-programs to large-scale, coarse-grain software engine- ing problems: contractual imperative program synthesis and structured p- gram synthesis. These applications constitute an exemplary justi?cation of the framework. Also, in and of themselves, these approaches to synthesis should be interesting forresearchers working in the target problem domains.
Distributed Collaborative Machine Learning (DCML) has emerged in artificial intelligence-empowered edge computing environments, such as the Industrial Internet of Things (IIoT), to process tremendous data generated by...
详细信息
Distributed Collaborative Machine Learning (DCML) has emerged in artificial intelligence-empowered edge computing environments, such as the Industrial Internet of Things (IIoT), to process tremendous data generated by smart devices. However, parallel DCML frameworks require resource-constrained devices to update the entire Deep Neural Network (DNN) models and are vulnerable to reconstruction attacks. Concurrently, the serial DCML frameworks suffer from training efficiency problems due to their serial training nature. In this paper, we propose a Model Pruning-enabled Federated Split Learning framework (MP-FSL) to reduce resource consumption with a secure and efficient training scheme. Specifically, MP-FSL compresses DNN models by adaptive channel pruning and splits each compressed model into two parts that are assigned to the client and the server. Meanwhile, MP-FSL adopts a novel aggregation algorithm to aggregate the pruned heterogeneous models. We implement MP-FSL with a real FL platform to evaluate its performance. The experimental results show that MP-FSL outperforms the state-of-the-art frameworks in model accuracy by up to 1.35%, while concurrently reducing storage and computational resource consumption by up to 32.2% and 26.73%, respectively. These results demonstrate that MP-FSL is a comprehensive solution to the challenges faced by DCML, with superior performance in both reduced resource consumption and enhanced model performance.
This book constitutes the refereed proceedings of the 10th International Conference on softwareengineering and Formal Methods, SEFM 2012, held in Thessaloniki, Greece, in October 2012. The 19 revised research papers ...
详细信息
ISBN:
(数字)9783642338267
ISBN:
(纸本)9783642338250
This book constitutes the refereed proceedings of the 10th International Conference on softwareengineering and Formal Methods, SEFM 2012, held in Thessaloniki, Greece, in October 2012. The 19 revised research papers presented together with 3 short papers, 2 tool papers, and 2 invited talks were carefully reviewed and selected from 98 full submissions. The SEFM conference aspires to advance the state-of-the-art in formal methods, to enhance their scalability and usability with regards to their application in the software industry and to promote their integration with practical engineering methods.
Since the first international conference on urban air quality, held at the University ofHertfordshire in 1996, significant advances have taken place in the field of urban air pollution. In addition to the scientific a...
详细信息
ISBN:
(数字)9789401009324
ISBN:
(纸本)9780792366768;9789401037969
Since the first international conference on urban air quality, held at the University ofHertfordshire in 1996, significant advances have taken place in the field of urban air pollution. In addition to the scientific advances in the measurement, modelling and management of urban air quality, significant progress has been achieved in relation to the establishment of major frameworks to ensure a more effective mechanism for international collaboration. Two such frameworks are SATURN (Studying Atmospheric Pollution in Urban Areas) and TRAPOS (Optimisation of Modelling Methods for Traffic Pollution in Streets). In response to such advances, the second international conference was held at the Technical University of Madrid in March 1999 with active participation of SATURN and TRAPOS investigators. The organisation of the conference was headed by the Institute of Physics in collaboration with the Technical University of Madrid and the University of Hertfordshire. The support of IUAPPA and AWMA ensured a truly worldwide promotion and participation. The meeting attracted 140 scientists from 26 different countries establishing it as a major forum for exchanging and discussing the latest research fmdings in this field.
暂无评论