The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)

Automating Electronic Circuit Design with Deep RL w/ Karim Beguir - #365

Apr 13, 2020

35:04

Snipd AI

Karim Beguir, Co-Founder and CEO of InstaDeep, discusses their new offering DeepPCB, an end-to-end platform for automated circuit board design. They talk about the challenges with auto-routers, defining circuit board complexity, the differences between reinforcement learning for games and circuit design, and their spotlight paper from NeurIPS.

AI Summary

AI Chapters

Episode notes

Podcast summary created with Snipd AI

Quick takeaways

InstaDeep has developed DeepPCB, an AI system powered by deep reinforcement learning, that can automate the complex task of routing chips on printed circuit boards (PCBs).

Deep reinforcement learning allows the system to continuously improve its performance by learning from experience, playing games, and receiving rewards based on completing designs and satisfying design rules.

Deep dives

Overview of InstaDeep's Progress

InstaDeep, a decision-making AI startup, has made significant progress in three areas. Firstly, they have released innovative products in decision-making AI. Secondly, they have been active in research, publishing original pieces and even receiving a spotlight presentation at NURIPS with Google DeepMind. Lastly, InstaDeep has been involved in community outreach, organizing major events in Africa and helping aspiring AI talents find opportunities.

Introduction

2min

DeepPCB: Automating Chip Routing on PCBs

18min

Automating Electronic Circuit Design using Deep RL

7min

Transfer Learning for Electronic Circuit Design

3min

Spotlight Presentation at NURPS and Collaboration with DeepMind

2min

Exploring the Paper Alpha NPI and Its Results

3min

Today we’re joined by return guest Karim Beguir, Co-Founder and CEO of InstaDeep. In our conversation, we chat with Karim about InstaDeep’s new offering, DeepPCB, an end-to-end platform for automated circuit board design. We discuss challenges and problems with some of the original iterations of auto-routers, how Karim defines circuit board “complexity,” the differences between reinforcement learning being used for games and in this use case, and their spotlight paper from NeurIPS.

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode

Save any
moment

Hear something you like? Tap your headphones to save it with AI-generated key takeaways

Share
& Export