The Valluvan app is a language solution for native Tamil speakers. The system emphasizes the recognition of name boards, translation, and speech output to enhance communication and access to information. The app utili...
详细信息
The main purpose of the project is to help the farmers who are not aware of the crop which suits their soil quality, soil nutrients, soil composition, the amount of water and its nutrients the crop needs and to identi...
详细信息
Good quality learning data makes it possible to create better models using machinelearning. Today, solutions based on artificial neural networks and deep learning are gaining more popularity. Unfortunately, these sol...
详细信息
The need for accurate time series forecasting has questioned the potential of Federated learning (FL) in solving regression problems with privacy-preserving and collaborative prognosis requirements. While recent Machi...
详细信息
ISBN:
(数字)9783031342042
ISBN:
(纸本)9783031342035;9783031342042
The need for accurate time series forecasting has questioned the potential of Federated learning (FL) in solving regression problems with privacy-preserving and collaborative prognosis requirements. While recent machinelearning (ML) studies have shown accurate predictions in time series forecasting using functional principal component analysis, the potential of integrating this approach with FL has not been previously evaluated. This paper depicts the potential of combining functional time series regression with FL through the implementation of a Functional Multilayer Perceptron (FMLP). Experimental results on one of the most innovative industrial maintenance strategies, Predictive Maintenance (PM), demonstrate that the integration of FMLP with the well-known Federated Averaging (FedAvg) algorithm achieves accurate time series forecasting while preserving data privacy. These results, obtained using NASA C-MAPSS datasets, outperformed traditional ML and Deep learning (DL) approaches in estimating the Remaining Useful Life (RUL) of aircraft components.
The automatic image colorization technique has garnered a lot of attention over the past ten years for a variety of applications, including the restoration of old or damaged photos. Owing to many level of independence...
详细信息
Deep Neural Networks (DNNs) are being adopted as components in software systems. Creating and specializing DNNs from scratch has grown increasingly difficult as state-of-the-art architectures grow more complex. Follow...
详细信息
ISBN:
(纸本)9781665457019
Deep Neural Networks (DNNs) are being adopted as components in software systems. Creating and specializing DNNs from scratch has grown increasingly difficult as state-of-the-art architectures grow more complex. Following the path of traditional software engineering, machinelearning engineers have begun to reuse large-scale pre-trained models (PTMs) and fine-tune these models for downstream tasks. Prior works have studied reuse practices for traditional software packages to guide software engineers towards better package maintenance and dependency management. We lack a similar foundation of knowledge to guide behaviors in pre-trained model ecosystems. In this work, we present the first empirical investigation of PTM reuse. We interviewed 12 practitioners from the most popular PTM ecosystem, Hugging Face, to learn the practices and challenges of PTM reuse. From this data, we model the decision-making process for PTM reuse. Based on the identified practices, we describe useful attributes for model reuse, including provenance, reproducibility, and portability. Three challenges for PTM reuse are missing attributes, discrepancies between claimed and actual performance, and model risks. We substantiate these identified challenges with systematic measurements in the Hugging Face ecosystem. Our work informs future directions on optimizing deep learning ecosystems by automated measuring useful attributes and potential attacks, and envision future research on infrastructure and standardization for model registries.
machinelearning paradigms are evolved as general need algorithms for data science applications. Unfortunately, their performance is degraded when the quality of the dataset is not up to the mark. It does mean that th...
详细信息
machinelearning is increasingly crucial for predicting application performance, offering a black-box approach that does not require a deep understanding of the application internal workings. This method enables accur...
详细信息
Because of the increased market opening and massive data collection brought about by globalization, the need of maintaining control over customs procedures has increased. However, the integration and processing of cus...
详细信息
advances in surveillance cameras that have risen with the need for increased security and prevention of real- time crimes have become very common in public. Advanced analyses in machinelearning can detect crime from ...
详细信息
暂无评论