Using artificial neural networks for analog integrated circuit design automation. (2020)

• Record Type: Book
• Further Information: Note: João P.S. Rosa [and more].
• Other Names: Rosa, João P. S; Guerra, Daniel J. D; Horta, Nuno C. G; Martins, Ricardo (Ferreira Martins); Lourenço, Nuno C. C
• Contents: Intro -- Preface -- Contents -- Abbreviations -- List of Figures -- List of Tables -- 1 Introduction -- 1.1 Analog Integrated Circuit Design Automation -- 1.2 Analog IC Design Flow -- 1.3 Machine Learning and Analog IC Sizing and Placement -- 1.4 Conclusions -- References -- 2 Related Work -- 2.1 Existing Approaches to Analog IC Sizing Automation -- 2.1.1 Machine Learning Applied to Sizing Automation -- 2.2 Existing Approaches to Automatic Layout Generation -- 2.2.1 Layout Generation with Placement and Routing Constraints -- 2.2.2 Layout Migration and Retargeting 2.2.3 Layout Synthesis with Knowledge Mining -- 2.2.4 Analog Placement Constraints -- 2.3 Conclusions -- References -- 3 Overview of Artificial Neural Networks -- 3.1 Machine Learning Overview -- 3.1.1 Supervised or Unsupervised Learning -- 3.1.2 Batch or Online Learning -- 3.1.3 Instance-Based or Model-Based Learning -- 3.2 Challenges in Creating ML Systems -- 3.3 The Five Tribes of Machine Learning -- 3.3.1 Why Use ANNs for EDA -- 3.4 Neural Network Overview -- 3.4.1 Size of the Model -- 3.4.2 Activation Functions -- 3.4.3 How ANNs "Learn" -- 3.4.4 Optimizers 3.4.5 Early Stop, Regularization, and Dropout -- 3.5 Summary -- References -- 4 Using ANNs to Size Analog Integrated Circuits -- 4.1 Design Flow -- 4.2 Problem and Dataset Definition -- 4.3 Regression-Only Model -- 4.3.1 Training -- 4.3.2 Using the ANN for Circuit Sizing -- 4.4 Classification and Regression Model -- 4.4.1 Training -- 4.5 Test Case-Regression-Only Model -- 4.5.1 Single-Stage Amplifier with Voltage Combiners -- 4.5.2 Two-Stage Miller Amplifier -- 4.5.3 Test Case-Classification and Regression Model -- 4.6 Conclusions -- References 5 ANNs as an Alternative for Automatic Analog IC Placement -- 5.1 Layout Synthesis by Deep Learning -- 5.2 Development of an ANN Model -- 5.2.1 Circuit Used for Tests -- 5.2.2 Dataset Architecture -- 5.2.3 Neural Network Architecture -- 5.2.4 Preprocessing the Data -- 5.2.5 Metrics to Evaluate the Models -- 5.2.6 Training of the Network -- 5.3 Experimental Results -- 5.3.1 Single-Layer Perceptron -- 5.3.2 Multilayer Perceptron -- 5.3.3 Multiples Templates -- 5.3.4 Multiples Placement Output -- 5.4 Conclusion and Future Research Directions -- References
• Publisher Details: Cham : Springer
• Publication Date: 2020
• Extent: 1 online resource (117 pages)
• Subjects: 621.3815; Integrated circuits -- Computer-aided design; Analog integrated circuits -- Computer-aided design; Integrated circuits -- Computer-aided design; Electronic books
• Languages: English
• ISBNs: 9783030357436; 3030357430; 3030357422; 9783030357429; 9783030357443; 3030357449
• Related ISBNs: 9783030357429
• Notes: Note: Includes bibliographical references.; Note: Print version record.
• Access Rights: Legal Deposit; Only available on premises controlled by the deposit library and to one user at any one time; The Legal Deposit Libraries (Non-Print Works) Regulations (UK).
• Access Usage: Restricted: Printing from this resource is governed by The Legal Deposit Libraries (Non-Print Works) Regulations (UK) and UK copyright law currently in force.