The Local Maximum

Max Sklar starts off the Local Maximum podcast. He goes over the purpose of the podcast, and the push for 10 episodes. Aaron Bell interviews Max on Bayesian vs Frequentist Statistics. Mathematical reverends, German Enigma's, and Tampered Dice.

Ep. 0 - The great beginning, and how to Update Your Beliefs

In the rest of my discussion with Aaron, we discuss the how Bayes rule can apply to news items like the Hawaii ICBM Scare, the murder of DNC Staffer Seth Rich, Medical Diagnosis, Conspiracy Theories, theology, and politics.

Ep. 1 - Bayesian Inference Applied to the Hawaii ICBM Scare

In my first interview, I speak to Google engineer Maryam Aly who worked on Foursquare's Internationalization Stack for several years. This includes making sure that the foursquare city guide and swarm apps work all around the world. Sometimes, a simple translation isn't enough - we had to deal with the quirks of each language, and puns and cultural references that don't carry through.

Local Maximum

Ep. 2 - Maryam Aly on Building a Global User Base: Translation, Localization, and I18n

I have a discussion with Stephanie Yang about how we used machine learning and sentiment analysis to build Foursquare's critically acclaimed venue rating system. Some anecdotes and what we learn along the way.

Ep. 3 - The Perfect 10, Foursquare's rating system

IN WAS NDE BESN OF NIZES IN WAS NDE WORSN OF NIZES IN WAS NDE ACE OF WISTOZ IN WAS NDE ACE OF FOOLISDGESS IN WAS NDE EPOUD OF BELIEF IN WAS NDE EPOUD OF IGURETMLINY IN WAS NDE SEASOG OF LICDN IN WAS NDE SEASOG OF TARQGESS IN WAS NDE SPRIGC OF DOPE IN WAS NDE WIGNER OF TESPAIR WE DAT EVERYNDIGC BEFORE MS WE DAT GONDIGC BEFORE MS WE WERE ALL COIGC TIREUN NO DEAVEG WE WERE ALL COIGC TIREUN NDE ONDER WAY (I'll explain!)

Ep. 4 - Codebreaking, Bizarro Harry Potter, and the Proper way to Gerrymander

Convolutional nets are great for image classification... if this were 2016. But it's 2018 and Canada's greatest neural networker Geoff Hinton has some new ideas, namely capsule networks. Capsule nets are a completely new type of neural net architecture designed to do image classification on far fewer training cases than convolutional nets, and they're posting results that are competitive with much more mature technologies.

In this episode, we'll give a light conceptual introduction to capsule nets and get geared up for a future episode that will do a deeper technical dive.

A Conceptual Introduction to Capsule Networks

Last episode we talked conceptually about capsule networks, the latest and greatest computer vision innovation to come out of Geoff Hinton's lab. This week we're getting a little more into the technical details, for those of you ready to have your mind stretched.

A Technical Introduction to Capsule Networks

A huge part of the ascent of deep learning in the last few years is related to advances in computer hardware that makes it possible to do the computational heavy lifting required to build models with thousands or even millions of tunable parameters. This week we'll pretend to be electrical engineers and talk about how modern machine learning is enabled by hardware.

CPUs, GPUs, TPUs: Hardware for Deep Learning

Join Max Sklar and guests in this podcast about AI, technology, and society. Email localmaxradio@gmail.com for questions and feedback.

Show notes page on the website: http://www.localmaxradio.com