Optimizing Hyperparameters for Machine Learning Algorithms in Production PDF Download
Are you looking for read ebook online? Search for your book and save it on your Kindle device, PC, phones or tablets. Download Optimizing Hyperparameters for Machine Learning Algorithms in Production PDF full book. Access full book title Optimizing Hyperparameters for Machine Learning Algorithms in Production by Jonathan Krauß. Download full books in PDF and EPUB format.
Author: Jonathan Krauß Publisher: Apprimus Wissenschaftsverlag ISBN: 3985550743 Category : Technology & Engineering Languages : en Pages : 258
Book Description
Machine learning (ML) offers the potential to train data-based models and therefore to extract knowledge from data. Due to an increase in networking and digitalization, data and consequently the application of ML are growing in production. The creation of ML models includes several tasks that need to be conducted within data integration, data preparation, modeling, and deployment. One key design decision in this context is the selection of the hyperparameters of an ML algorithm – regardless of whether this task is conducted manually by a data scientist or automatically by an AutoML system. Therefore, data scientists and AutoML systems rely on hyperparameter optimization (HPO) techniques: algorithms that automatically identify good hyperparameters for ML algorithms. The selection of the HPO technique is of great relevance, since it can improve the final performance of an ML model by up to 62 % and reduce its errors by up to 95 %, compared to computing with default values. As the selection of the HPO technique depends on different domain-specific influences, it becomes more and more popular to use decision support systems to facilitate this selection. Since no approach exists, which covers the requirements from the production domain, the main research question of this thesis was: Can a decision support system be developed that supports in the selecting of HPO techniques in the production domain?
Author: Jonathan Krauß Publisher: Apprimus Wissenschaftsverlag ISBN: 3985550743 Category : Technology & Engineering Languages : en Pages : 258
Book Description
Machine learning (ML) offers the potential to train data-based models and therefore to extract knowledge from data. Due to an increase in networking and digitalization, data and consequently the application of ML are growing in production. The creation of ML models includes several tasks that need to be conducted within data integration, data preparation, modeling, and deployment. One key design decision in this context is the selection of the hyperparameters of an ML algorithm – regardless of whether this task is conducted manually by a data scientist or automatically by an AutoML system. Therefore, data scientists and AutoML systems rely on hyperparameter optimization (HPO) techniques: algorithms that automatically identify good hyperparameters for ML algorithms. The selection of the HPO technique is of great relevance, since it can improve the final performance of an ML model by up to 62 % and reduce its errors by up to 95 %, compared to computing with default values. As the selection of the HPO technique depends on different domain-specific influences, it becomes more and more popular to use decision support systems to facilitate this selection. Since no approach exists, which covers the requirements from the production domain, the main research question of this thesis was: Can a decision support system be developed that supports in the selecting of HPO techniques in the production domain?
Author: DuoDuo Ying Publisher: ISBN: Category : Languages : en Pages : 0
Book Description
Deep learning algorithms are increasingly popular for complex prediction and classification tasks, and hyperparameter configurations play an important role in algorithm performance. However, the best hyperparameter tuning strategy still remain unresolved. While grid search and random search can be used to detect better hyperparameters, they are costly for big deep learning algorithms and may not produce the optimal result. Bayesian Optimization balancing the exploration and exploitation trade-off shows significant improvement over grid search and random search in both efficiency and accuracy, but the algorithm makes computation on the entire domain, which can be still costly especially in higher dimension settings. In this paper, we propose a space adjustment algorithm selecting top percent points at each iteration that can be incorporated in additional to Bayesian Optimization framework to further reduce experimental cost and improve optimization efficiency. We show our algorithm's adaptable nature to the response surface of hyperparameter configuration space. We demonstrate our algorithm's outstanding performance compared with Efficient Global Optimization through a variety of test functions and an application to machine learning datasets.
Author: Samantha Corinne Sanders Publisher: ISBN: Category : Electronic dissertations Languages : en Pages : 33
Book Description
One of the challenges of data mining is finding hyper-parameters for a learning algorithm that will produce the best model for a given dataset. Hyper-parameter optimization automates this process, but it can still take significant time. It has been found that hyperparameter optimization does not always result in induced models with significant improvement over default hyper-parameters, yet no systematic analysis of the role of hyper-parameter optimization in machine learning has been conducted. We propose the use of meta-learning to inform the decision to optimize hyper-parameters based on whether default hyper-parameter performance can be surpassed in a given amount of time. We will build a base of metaknowledge, through a series of experiments, to build predictive models that will assist in the decision process.
Author: Zekâi Şen Publisher: Springer Nature ISBN: 9819960622 Category : Technology & Engineering Languages : en Pages : 824
Book Description
This book comprises the proceedings of the 12th International Symposium on Intelligent Manufacturing and Service Systems 2023. The contents of this volume focus on recent technological advances in the field of artificial intelligence in manufacturing & service systems including machine learning, autonomous control, bioinformatics, human-artificial intelligence interaction, digital twin, robotic systems, sybersecurity, etc. This volume will prove a valuable resource for those in academia and industry.
Author: Kim Phuc Tran Publisher: Springer Nature ISBN: 3030838196 Category : Technology & Engineering Languages : en Pages : 270
Book Description
This book introduces the latest research on advanced control charts and new machine learning approaches to detect abnormalities in the smart manufacturing process. By approaching anomaly detection using both statistics and machine learning, the book promotes interdisciplinary cooperation between the research communities, to jointly develop new anomaly detection approaches that are more suitable for the 4.0 Industrial Revolution. The book provides ready-to-use algorithms and parameter sheets, enabling readers to design advanced control charts and machine learning-based approaches for anomaly detection in manufacturing. Case studies are introduced in each chapter to help practitioners easily apply these tools to real-world manufacturing processes. The book is of interest to researchers, industrial experts, and postgraduate students in the fields of industrial engineering, automation, statistical learning, and manufacturing industries.
Author: Frank Hutter Publisher: Springer ISBN: 3030053180 Category : Computers Languages : en Pages : 223
Book Description
This open access book presents the first comprehensive overview of general methods in Automated Machine Learning (AutoML), collects descriptions of existing systems based on these methods, and discusses the first series of international challenges of AutoML systems. The recent success of commercial ML applications and the rapid growth of the field has created a high demand for off-the-shelf ML methods that can be used easily and without expert knowledge. However, many of the recent machine learning successes crucially rely on human experts, who manually select appropriate ML architectures (deep learning architectures or more traditional ML workflows) and their hyperparameters. To overcome this problem, the field of AutoML targets a progressive automation of machine learning, based on principles from optimization and machine learning itself. This book serves as a point of entry into this quickly-developing field for researchers and advanced students alike, as well as providing a reference for practitioners aiming to use AutoML in their work.
Author: J. Andersson Publisher: IOS Press ISBN: 1643685112 Category : Technology & Engineering Languages : en Pages : 750
Book Description
Collaboration between those working in product development and production is essential for successful product realization. The Swedish Production Academy (SPA) was founded in 2006 with the aim of driving and developing production research and higher education in Sweden, and increasing national cooperation in research and education within the area of production. This book presents the proceedings of SPS2024, the 11th Swedish Production Symposium, held from 23 to 26 April 2024 in Trollhättan, Sweden. The conference provided a platform for SPA members, as well as for professionals from industry and academia interested in production research and education from around the world, to share insights and ideas. The title and overarching theme of SPS2024 was Sustainable Production through Advanced Manufacturing, Intelligent Automation and Work Integrated Learning, and the conference emphasized stakeholder value, the societal role of industry, worker wellbeing, and environmental sustainability, in alignment with the European Commission's vision for the future of manufacturing. The 59 papers included here were accepted for publication and presentation at the symposium after a thorough review process. They are divided into 6 sections reflecting the thematic areas of the conference, which were: sustainable manufacturing, smart production and automation, digitalization for efficient product realization, circular production, industrial transformation for sustainability, and the integration of education and research. Highlighting the latest developments and advances in automation and sustainable production, the book will be of interest to all those working in the field.
Author: Anjan Kumar Chatterjee Publisher: CRC Press ISBN: 1000475646 Category : Technology & Engineering Languages : en Pages : 493
Book Description
This book captures the path of digital transformation that the cement enterprises are adopting progressively to elevate themselves to ‘Industry 4.0’ level. Digital innovations-based Internet of Things (IoT) and Artificial Intelligence (AI) are pertinent technologies for the cement enterprises as the manufacturing processes operate at very large scales with multiple inputs, outputs, and variables, resulting in the essentiality of big data management. Featuring contributions from cement industries worldwide, it covers various aspects of cement manufacturing from IoT, machine learning and data analytics perspective. It further discusses implementation of digital solutions in cement process and plants through case studies. Features: Present an up-to-date, consolidated view on modern cement manufacturing technology, applying new systems. Provides narration of complexity and variables in modern cement plants and processes. Discusses evolution of automation and computerization for the manufacturing processes. Covers application of ERP techniques to cement enterprises. Includes data-driven approaches for energy, environment, and quality management. This book aims at researchers and industry professionals involved in cement manufacturing, cement machinery and system suppliers, chemical engineering, process engineering, industrial engineering, and chemistry.
Author: Thomas V Joseph Publisher: BPB Publications ISBN: 9355516495 Category : Computers Languages : en Pages : 402
Book Description
Architecting AI: Design patterns for building deep learning products KEY FEATURES ● Master foundational concepts in design patterns of deep learning. ● Benefit from practical insights shared by an industry professional. ● Learn to build data products using deep learning. DESCRIPTION Design Patterns of Deep Learning with TensorFlow is your comprehensive guide to learning deep learning from a design pattern perspective. In this book, we explore deep learning within the context of building hyper-personalization models, exploring its applications across various industries and scenarios. It starts by showing how deep learning enhances retail through customer segmentation and data analysis. You will learn neural networks, computer vision with CNNs, and NLP for analyzing customer behavior. This book addresses challenges like uneven data and optimizing models with techniques like backpropagation, hyperparameter tuning, and transfer learning. Finally, it covers setting up data pipelines and deploying your system. With practical tips and actionable advice, this book equips readers with the skills and strategies needed to thrive in today's competitive AI landscape. By the end of this book, you will be equipped with the knowledge and practical skills to build and deploy deep learning-powered hyper-personalization systems that deliver exceptional customer experiences. WHAT YOU WILL LEARN ● Understand about hyper-personalized AI models for tailored user experiences. ● Design principles of computer vision and NLP models. ● Inner working of transformers equipping readers to understand the intricacies of generative AI and large language models (LLMs) like ChatGPT. ● To get the best out of deep learning models through hyperparameter tuning and transfer learning. ● Learn how to build deployment pipelines to serve models into production environments seamlessly. WHO THIS BOOK IS FOR This book caters to both beginners and experienced practitioners in the field of data science and Machine Learning. Through practical examples, it simplifies complex ideas, linking them to design patterns. TABLE OF CONTENTS 1. Customer Hyper-personalization 2. Introduction to Design Patterns and Neural Networks 3. Design Patterns in Visual Representation Learning 4. Design Patterns for Non-Visual Representation Learning 5. Design Patterns for Transformers 6. Data Distribution Challenges and Strategies 7. Model Training Philosophies 8. Hyperparameter Tuning 9. Transfer Learning 10. Setting Up Data and Deployment Pipelines
Author: Apoorva S. Shastri Publisher: Springer Nature ISBN: 9819974569 Category : Computers Languages : en Pages : 175
Book Description
This book aims to provide a collection of state-of-the-art scientific and technical research papers related to machine learning-based algorithms in the field of optimization and engineering design. The theoretical and practical development for numerous engineering applications such as smart homes, ICT-based irrigation systems, academic success prediction, future agro-industry for crop production, disease classification in plants, dental problems and solutions, loan eligibility processing, etc., and their implementation with several case studies and literature reviews are included as self-contained chapters. Additionally, the book intends to highlight the importance of study and effectiveness in addressing the time and space complexity of problems and enhancing accuracy, analysis, and validations for different practical applications by acknowledging the state-of-the-art literature survey. The book targets a larger audience by exploring multidisciplinary research directions such as computer vision, machine learning, artificial intelligence, modified/newly developed machine learning algorithms, etc., to enhance engineering design applications for society. State-of-the-art research work with illustrations and exercises along with pseudo-code has been provided here.
Author: Jonathan Krauß
Author: Jonathan Krauß
Author: DuoDuo Ying
Author: Samantha Corinne Sanders
Author: Zekâi Şen
Author: Kim Phuc Tran
Author: Frank Hutter
Author: J. Andersson
Author: Anjan Kumar Chatterjee
Author: Thomas V Joseph
Author: Apoorva S. Shastri