Filters








83 Hits in 5.0 sec

Amazon SageMaker Model Monitor: A System for Real-Time Insights into Deployed Machine Learning Models [article]

David Nigenda, Zohar Karnin, Muhammad Bilal Zafar, Raghu Ramesha, Alan Tan, Michele Donini, Krishnaram Kenthapadi
2022 arXiv   pre-print
We present Amazon SageMaker Model Monitor, a fully managed service that continuously monitors the quality of machine learning models hosted on Amazon SageMaker.  ...  With the increasing adoption of machine learning (ML) models and systems in high-stakes settings across different industries, guaranteeing a model's performance after deployment has become crucial.  ...  Acknowledgments The authors would like to thank other members of the Amazon AWS AI team for their collaboration during the development and deployment of Amazon Sage-Maker Model Monitor, and Jay Casteel  ... 
arXiv:2111.13657v3 fatcat:x37pz6epvjcadly6537ohq257e

Amazon SageMaker Debugger: A System for Real-Time Insights into Machine Learning Model Training

Nathalie Rauschmayr, Vikas Kumar, Rahul Huilgol, Andrea Olgiati, Satadal Bhattacharjee, Nihal Harish, Vandana Kannan, Amol Lele, Anirudh Acharya, Jared Nielsen, Lakshmi Ramakrishnan, Ishan Bhatt (+13 others)
2021 Conference on Machine Learning and Systems  
model training in real-time.  ...  We present Amazon SageMaker Debugger, a machine learning feature that automatically identifies and stops underperforming training jobs.  ...  Amazon SageMaker Amazon SageMaker is a fully managed service provided as part of Amazon Web Services (AWS) that enables data scientists and developers to build, train, and deploy ML models in the cloud  ... 
dblp:conf/mlsys/RauschmayrKHOBH21 fatcat:5otljq6wrnhllf7zh7m7hbtove

Model Monitoring in Practice

Krishnaram Kenthapadi, Himabindu Lakkaraju, Pradeep Natarajan, Mehrnoosh Sameki
2022 Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining  
However, there is relatively less attention on the need for monitoring machine learning (ML) models once they are deployed and the associated research challenges.  ...  developer perspectives, and provide a roadmap for thinking about model monitoring in practice.  ...  and lessons learned from deploying model monitoring tools for several web-scale AI/ML applications.  ... 
doi:10.1145/3534678.3542617 fatcat:qzx4q44psng6djtclh6dmmpc3q

Amazon SageMaker Clarify: Machine Learning Bias Detection and Explainability in the Cloud

Michaela Hardt, Xiaoguang Chen, Xiaoyi Cheng, Michele Donini, Jason Gelman, Satish Gollaprolu, John He, Pedro Larroy, Xinyu Liu, Nick McCarthy, Ashish Rathi, Scott Rees (+9 others)
2021 arXiv   pre-print
We present Amazon SageMaker Clarify, an explainability feature for Amazon SageMaker that launched in December 2020, providing insights into data and ML models by identifying biases and explaining predictions  ...  It is deeply integrated into Amazon SageMaker, a fully managed service that enables data scientists and developers to build, train, and deploy ML models at any scale.  ...  We thank Edi Razum, Luuk Figdor, Hasan Poonawala, the DFL and Zopa for contributions to the customer use case, and Cédric Archambeau and the anonymous reviewers for their valuable feedback.  ... 
arXiv:2109.03285v1 fatcat:k5rtnrvs3zarxavjxqdl4p4hda

A Human-Centric Take on Model Monitoring [article]

Murtuza N Shergadwala, Himabindu Lakkaraju, Krishnaram Kenthapadi
2022 arXiv   pre-print
However, there is a research gap in understanding the human-centric needs and challenges of monitoring machine learning (ML) models once they are deployed.  ...  We identified various human-centric challenges and requirements for model monitoring in real-world applications.  ...  Acknowledgments We are thankful to the interviewees for their detailed responses to the questions. We also thank Joshua Rubin and Lea Genuit for their feedback and help with the analysis.  ... 
arXiv:2206.02868v2 fatcat:ufq3f4cxobaylacjfvanfhhx7y

An Automated Data Engineering Pipeline for Anomaly Detection of IoT Sensor Data [article]

Xinze Li, Baixi Zou
2021 arXiv   pre-print
The process involves the use of IoT sensors, Raspberry Pis, Amazon Web Services (AWS) and multiple machine learning techniques with the intent to identify anomalous cases for the smart home security system  ...  With data analytics and the use of machine learning/deep learning, it is made possible to learn the underlying patterns and make decisions based on what was learned from massive data generated from IoT  ...  Li Yang from OC2 Lab, Department of Electrical and Computer Engineering, Western University, for their support and help throughout this work.  ... 
arXiv:2109.13828v1 fatcat:ste4fwe35jasxcxln4dlxqze5i

Serverless Model Serving for Data Science [article]

Yuncheng Wu, Tien Tuan Anh Dinh, Guoyu Hu, Meihui Zhang, Yeow Meng Chee, Beng Chin Ooi
2021 arXiv   pre-print
Machine learning (ML) is an important part of modern data science applications.  ...  We conduct a comprehensive evaluation of the performance and cost of serverless against other model serving systems on two clouds: Amazon Web Service (AWS) and Google Cloud Platform (GCP).  ...  Machine learning service. Most cloud providers offer fully managed services for scientists to build, train, and deploy ML models easily.  ... 
arXiv:2103.02958v1 fatcat:mdfnswimkfce3dxan6zjrliyfm

Machine Learning Operations (MLOps): Overview, Definition, and Architecture [article]

Dominik Kreuzberger, Niklas Kühl, Sebastian Hirschl
2022 arXiv   pre-print
The final goal of all industrial machine learning (ML) projects is to develop ML products and rapidly bring them into production.  ...  However, MLOps is still a vague term and its consequences for researchers and professionals are ambiguous.  ...  Furthermore, the academic space has focused intensively on machine learning model building and benchmarking, but too little on operating complex machine learning systems in real-world scenarios.  ... 
arXiv:2205.02302v3 fatcat:bzermjilzvh35bpmqnv2yfzvh4

Health Monitoring IoT Device with Risk Prediction using Cloud Computing and Machine Learning

Anindya Das, Zannatun Nayeem, Abu Saleh Faysal, Fardoush Hassan Himu, Tanvinur Rahman Siam
2021 2021 National Computing Colleges Conference (NCCC)  
First, we utilize the Amazon Sagemaker the machine learning tools for the Amazon Web Services [29] . The training data on about (80 percent) will be utilized during the model preparing circle.  ...  As our system is mainly a prediction based system we needed a cloud service that deals very efficiently with machine learning models.  ... 
doi:10.1109/nccc49330.2021.9428798 fatcat:xgbv4wh6cfcstfq6xmoule7sz4

Automated Data Validation in Machine Learning Systems

Felix Biessmann, Jacek Golebiowski, Tammo Rukat, Dustin Lange, Philipp Schmidt
2021 IEEE Data Engineering Bulletin  
Machine Learning (ML) algorithms are a standard component of modern software systems.  ...  Machine Learning (ML) technology has become a standard component in modern software systems.  ...  Another example is SageMaker Clarify 15 , an explainability feature for Amazon SageMaker that provides insights into data and ML models by identifying biases and explaining predictions.  ... 
dblp:journals/debu/BiessmannGRL021 fatcat:x3fugemqhjgsramr26th3eu2oe

Practices and Infrastructures for Machine Learning Systems: An Interview Study in Finnish Organizations

Dennis Muiruri, Lucy Ellen Lwakatare, Jukka K. Nurminen, Tommi Mikkonen
2022 Computer  
The research method was selected to gain a deep understanding of the enacted practices and tool support for ML systems in real-world C O M P U T E R W W W . C O M P U T E R .  ...  Docker images created from the CI system are (automatically) deployed to a staging environment for additional tests before deployment to production.  ... 
doi:10.1109/mc.2022.3161161 fatcat:ib4xtmw3njb4lmvktihfylzbn4

Serverless Architectures Review, Future Trend and the Solutions to Open Problems

Manoj Kumar
2019 American Journal of Software Engineering  
Also provides comparative analysis on available serverless architectures for the most common use cases within cloud provider's environment.  ...  Lately, the popularity and adoption of serverless computing or Function-as-a-Service have been grown substantially, and it emerges as a better way to manage cost, reliability, availability, and scalability  ...  Cloud Dataflow: managed service for transforming and processing real-time data stream or in batch. Intelligence Amazon Machine Learning: service to perform predictions in real-time at scale.  ... 
doi:10.12691/ajse-6-1-1 fatcat:j7ufufymdrf2tbfedjryi2hq24

Embedded AI-Based Digi-Healthcare

Zarlish Ashfaq, Rafia Mumtaz, Abdur Rafay, Syed Mohammad Hassan Zaidi, Hadia Saleem, Sadaf Mumtaz, Adnan Shahid, Eli De Poorter, Ingrid Moerman
2022 Applied Sciences  
The system employs edge computing to perform multiple functionalities including health status inference using a Machine Learning (ML) model which makes predictions on real-time data, alert notifications  ...  A web-based application is developed for the depiction of raw data and ML results and to provide a direct communication channel between the patient and the doctor.  ...  Model Deployment After selecting and optimizing the KNNs model, the next step was to deploy the model at the edge for real-time prediction and monitoring of patients.  ... 
doi:10.3390/app12010519 fatcat:4nprmgi5qrdcllsjild6roexni

Dynamic Space-Time Scheduling for GPU Inference [article]

Paras Jain, Xiangxi Mo, Ajay Jain, Harikaran Subbaraj, Rehan Sohail Durrani, Alexey Tumanov, Joseph Gonzalez, Ion Stoica
2018 arXiv   pre-print
Our preliminary prototype of a dynamic space-time scheduler demonstrates a 3.23x floating-point throughput increase over space-only multiplexing and a 7.73x increase over time-only multiplexing for convolutions  ...  In this paper, we explore several techniques to leverage both temporal and spatial multiplexing to improve GPU utilization for deep learning inference workloads.  ...  Acknowledgements We thank Koushik Sen, Eyal Sela, Zongheng Yang, Anjali Shankar and Daniel Crankshaw for their insightful feedback and edits.  ... 
arXiv:1901.00041v1 fatcat:hzxlziaftvbypnwtls5a4uqk6e

MLOps: A Taxonomy and a Methodology

Matteo Testi, Matteo Ballabio, Emanuele Frontoni, Giulio Iannello, Sara Moccia, Paolo Soda, Gennaro Vessio
2022 IEEE Access  
Learning (ML).  ...  ML Operations (MLOps) represents an effective strategy for bringing ML models from academic resources to useful tools for solving problems in the corporate world.  ...  ML-BASED SOFTWARE SYSTEMS Machine Learning is becoming the primary approach to solving real-world problems.  ... 
doi:10.1109/access.2022.3181730 fatcat:4gokhtootvacjc2xjgeqkhrwhq
« Previous Showing results 1 — 15 out of 83 results