YoVDO

Cedar: A Composable ML Accelerated Analog Circuit Simulator

Offered By: The Julia Programming Language via YouTube

Tags

Julia Courses Machine Learning Courses

Course Description

Overview

Explore a groundbreaking conference talk from JuliaCon 2021 that introduces Cedar, a novel composable SPICE simulator written entirely in Julia for analog circuit simulation. Discover how this innovative tool aims to revolutionize technical computing in circuit design, matching it with modern standards. Learn about the transformation of circuit models into equations, the use of Symbolics.jl in circuit design, and the potential for differentiable circuit optimization. Delve into the challenges of surrogate models and Continuous-Time Echo State Networks in circuit simulation. Understand how machine learning is reshaping circuit design and how Julia Computing is bringing advanced digital community practices to analog circuits. Gain insights into the current state of analog design, its drawbacks, and the promising future offered by tools like Cedar and Xyce.jl. This 26-minute presentation covers a wide range of topics, from the basics of circuit simulation to cutting-edge approaches in analog circuit design and optimization.

Syllabus

Welcome!.
Agenda.
Technical computing is behind the times.
We aim at matching technical computing with the times.
Three simulation products from Julia Computing.
JuliaHub, a supercomputer at your fingertips.
A differential analog circuit simulator written in pure Julia.
How to transform circuit model into equation?.
Current state of parser and simulator.
Using Symbolics.jl in designing circuits.
Differentiable circuit optimization.
Our simulator is just a Julia library, so we can seamlessly use the whole Julia ecosystem.
ML is reshaping computing, including circuit design.
Black box and white box surrogatizaton of models.
Problem: surrogate models of circuits are stiff.
Challenges of dealing with Continuous-Time Echo State Networks (CTSN)?.
New approach to train oscillatory process (at the time of the talk, not published).
Analog design today is often the same as it was in the 1990s and has serious drawbacks.
Julia Computing brings to analog circuits what the digital community already does.
Be we want to do it even better.
Xyce.jl: wrapper for Xyce, a parallel circuit-simulator.
Summary.


Taught by

The Julia Programming Language

Related Courses

Julia Scientific Programming
University of Cape Town via Coursera
Julia for Beginners in Data Science
Coursera Project Network via Coursera
Linear Regression and Multiple Linear Regression in Julia
Coursera Project Network via Coursera
Decision Tree and Random Forest Classification using Julia
Coursera Project Network via Coursera
Logistic Regression for Classification using Julia
Coursera Project Network via Coursera