Early Bird Registration Now Open till November 30th. Save Up to $300 Today!


DesignCon 2019 Presentation Viewer

Purchase procecdings

Welcome to the DesignCon Presentation Store. Here you can view and download conference and/or show floor theater presentations before, during, and after the event. If you’re looking for a presentation from a specific session that you’re unable to find here, note that it’s likely because the presenter has not provided permission for external use or has not yet shared their presentation with us. Please check back after the event for a more complete catalogue of available presentations.

If you’d like to do a bulk download of all conference presentations or technical papers at once, please click here for conference presentations or click here for full technical papers. For sessions not included in the main conference, click here for Chiphead Theater presentations or click here for sponsored session presentations.

Boot Camp – Machine Learning & Artificial Intelligence for Hardware & Electronics Design

Christopher Cheng (Distinguished Technologist, Hewlett Packard Enterprise)

Paul Franzon (Cirrus Logic Distinguished Professor, North Carolina State University)

Madhavan Swaminathan (John Pippin Chair Professor, Georgia Institute of Technology)

YongJin Choi (Master Technologist, HP Enterprise)

Ting Zhu (Expert Engineer, HP Enterprise)

Seamus Brokaw (Software Engineer, Tektronix)

Majid Ahadi Dolatsara (Ph.D. Student, Georgia Institute of Technology, Atlanta)

Huan Yu (Ph.D. Student, Georgia Institute of Technology)

Hakki Mert Torun (Ph.D. Student, Georgia Institute of Technology)

Location: Great America 2

Date: Tuesday, January 29

Time: 9:00am - 4:30pm

Track: 15. Machine Learning for Microelectronics, Signaling & System Design

Session Type: Boot Camp

Vault Recording: TBD

Audience Level: All

A day long introduction for beginners who are interested in learning the basics of machine learning (ML) and artificial intelligence (AI) and their applications in hardware and electronics design. Participants will have hands-on opportunity to measure and train dynamic neural networks to illustrate its usefulness for complex equalizer modelling.

Please note: In order to take part in the hands-on instruction during this boot camp, you must bring your laptop and have TensorFlow and Keras software loaded and running.

Before the boot camp, go to the sites below and follow the instructions:

  • Download and install TensorFlow here
  • Download and install Keras here


Topics covered:

Introduction
We will start the day with a brief introduction to generative vs. discriminative models and their differences. From these models, we can consider examples where deploying machine learning/AI techniques can have big advantages and cases where they may not help.

Linear & logistic regression
This session introduces the basic concepts of linear and logistic regression in multidimensional spaces. It includes the mapping concept, and mentions regularized regression to address over-fitting. Participants will begin to learn how to use collection of input and output samples to implement generative models. Session ends with research examples applying regularized regression for nonlinear functions.

Machine learning for design optimization
The tutorial will cover the basics of machine learning based design optimization. Participants will learn how to use readily available tools and algorithms which can be integrated with their current simulation framework.

Artificial neural networks, regularization & gradient descent
We will move into the area of artificial neural networks. This is one of the most popular artificial intelligence engine. Concepts of input, output and hidden layer will be discussed. Activation functions and the overall operation of the ANN will also be discussed. Method to converge to optimal solutions for ANN through back propagation and gradient descent will be discussed. Undesirable results such as over/under fitting and their mitigation through regularization will be covered.

Bayesian surrogate models
Participants will be introduced to the important concept of Bayesian surrogate models to effectively model analog generative models

Recurrent neural networks
More advance neural network structures such as recurrent neural network will be discussed including their application in time series analysis for SI.

Advance topics in machine learning and AI (PCA and self-correcting models)
We will briefly touch on some advance topics and their application in hardware design. Examples will be using Principal component analysis for channel performance optimization, causal and structural inference for complex deep state space models such as hidden Markov and recurrent neural networks

Live demo of using test instruments as machine learning tools
The course will end with a hands on demo of using recurrent neural network to create a complex equalizing receiver model with test instruments. Measurements results will be compared with original IBIS-AMI models, any mismatch will be corrected through retraining of the recurrent neural network model (i.e. self-correcting model) without needing to know the IBIS or the receiver IP. Participants should bring their own Windows based laptop to run the program.



Takeaway


Intended Audience

None

Presentation Files

SLIDES_15_BootCampMachineLearningArtificial_Franzon2.pdf
SLIDES_15_MachineLearningMethodsinHigh_Wu.pdf