Minimizing Unwanted Biases (e.g., by Gender, Ethnicity) within ML Models

Here's a new blog post I wrote on how my team eliminates unwanted biases (e.g., by gender, ethnicity) from algorithms we've deployed in the recruitment sector.

Devising algorithms that stamp out unwanted biases without skimping on accuracy or performance adds time and effort to the machine learning model-design process. When algorithms can have a considerable social impact, as ours do in the human-resources space at GQR Global Markets, investing this time and effort is essential to ensuring equitable treatment of all people.

Vectors and Norms: Three New "ML Foundations" Videos

Three further videos in my Machine Learning Foundations series out today! Taking it up a dimension, these are on vector tensors and the norm functions we use to measure them:

• Video 5 in the series is "Vectors and Vector Transposition"
• Topic 6 is "Norms and Unit Vectors"
• And topic 7 is a quick one on "Basis, Orthogonal, and Orthonormal Vectors"

The YouTube playlist for the entire series, which will consist of 30+ hours of videos, is here.

The series is full of hands-on demos in NumPy, PyTorch, and TensorFlow, and all of the code is in GitHub here.

As usual, producer/editor Sangbin Lee did some fine work here.

"Hidden in Plain Sight" Podcast

I've been a guest on many wonderful podcasts, but this one on Hidden in Plain Sight easily takes the cake in terms of production value. I highly recommend tuning in for a candid take on the impacts of A.I. today and in the years to come: It’s titled Deep Learning Illustrated with Jon Krohn and can be found here.

Thank you Chad Grills and Lacey Peace for having me on the program. I've been thoroughly impressed by your operation from start to finish! 

Next Three Videos in the ML Foundations Series are Live

I published three new videos in my Machine Learning Foundations series today, including an introduction to tensors, with a focus on handling scalar tensors in PyTorch and TensorFlow (vectors and matrices coming up next).

I hope you enjoy them! The response to the first videos in the series -- released a fortnight ago -- has blown me over. I'm excited to continue to get more of these out on YouTube ASAP.

The playlist for the entire series, which will consist of 30+ hours of videos, is here.

The series is full of hands-on demos in Python and all of the code is in GitHub here.

Thanks to Sangbin Lee for continuing to be an outstanding producer and editor.

The "Machine Learning Foundations" video series is here!

Today I released the first videos of my popular Machine Learning Foundations series onto YouTube. It covers all the math and computer science needed to be an outstanding ML practitioner. 

Since May, I've been launching the content as live 3.5-hour webinars in the O'Reilly online learning platform and the response has been overwhelmingly positive, with classes garnering over 1200 registrations each and net promoter scores above 90%. In all, the series consists of eight of these 3.5-hour classes covering linear algebra, calculus, probability, statistics, algorithms, data structures, and optimization. (More detail on the series in GitHub here.)

As of this morning, thanks to superb editing from the talented Sangbin Lee, we're rolling out all of the content in the series as free YouTube videos.

A short welcome video is featured above, while the first true tutorial, on "What Linear Algebra Is", is featured below.

The playlist for the entire series, which will consist of 30+ hours of videos, is here.

Finally, the series is full of hands-on demos in Jupyter notebooks featuring Python, PyTorch, and TensorFlow code, and all of it is available via the GitHub link above.

More videos to come soon... 

New ODSC AI+ Training Platform

The brilliant folks behind the Open Data Science Conference (ODSC) are launching a new online training platform called AI+ and I'm delivering the inaugural class — Deep Learning with TensorFlow 2 and PyTorch — on July 29th!

New 2-Hour Tutorial on Machine Vision

New two-hour video introducing all the major machine vision concepts and (TensorFlow-based) applications, including:

• Convolutional neural networks (CNNs)

• Residual networks

• Object detection

• Image segmentation

• Transfer learning

• Capsule networks

We worked hard and long through filming, production, and editing to push the standard of what's possible in a software tutorial; I'm confident it's the best video I've ever been a part of.

All of the code demos are in Python and available in GitHub here.

DataScienceGO -- wow!

Thanks to the SuperDataScience team for providing me the opportunity to open up Day 2 of the tightly-run DataScienceGO conference on Sunday! For my talk on deep learning model architectures for natural language processing, there were over 400 advanced practitioners from 93 countries. They were attentive throughout the session and asked fabulously thoughtful questions during an extensive Q&A afterward.

DataScienceGO Virtual Conference

The DataScienceGO Virtual conference is this weekend and completely free!

Floored to be sharing the stage with DJ Patil and Emily Robinson, luminaries I’ve looked up to for years.

I’ll be doing an hour-long workshop on neural-network model architectures for natural language processing on Sunday.

Sign up is here.

SuperDataScience Podcast →

Was honored to be approached by Kirill to appear on his legendary SuperDataScience podcast and, wow, had a wonderful experience as a guest on the program. Thanks again, Kirill and team!

We covered the importance of data science in medicine and epidemiology, the role of data science in recruitment, testing your models for bias, what I think the future holds for deep learning and much more.

Machine Learning Foundations

I’m very excited to announce the beginning of a new journey called the Machine Learning Foundations series. As discussed in the video announcement above, this series will initially consist of eight 3.5-hour classes offered within the O’Reilly online learning platform from late May through early September:

1. May 28th — Intro to Linear Algebra

2. June 4th — Linear Algebra II: Matrix Operations

3. June 11th — Calculus I: Limits & Derivatives

4. June 25th —Calculus II: Partial Derivatives & Integrals

5. July 8th —Probability & Information Theory

6. July 23rd —Intro to Statistics

7. August 12th —Algorithms & Data Structures

8. early September — Optimization

Each class will feature:

• Rich, full-color illustrations

• Hands-on code demos in Python

• Fully worked-through pencil-and-paper questions and solutions

There are 605 seats available in each class. At the time of posting:

• The first two classes (on algebra) have fewer than ten seats each

• The second pair of classes (on calculus) have fewer than a hundred

• Registration for the fifth and sixth class is open and seats are filling up quickly.

As I have bandwidth, I will be publishing all of this content as free YouTube video tutorials, so if you’ve missed the classes, don’t worry! I will also probably be offering the classes again at some point, and eventually all of the content will be brought together neatly as a book.

You can read more about the Machine Learning Foundations series — including a detailed syllabus for each class and the developing body of open-source code — in GitHub here.

ODSC East Virtual Conference: So much fun, I couldn't resist singing

After my 3.5-hour technical lecture on Wednesday morning, I rewarded my 300 attendees’ remarkable attention spans with a cover of Willin’ by Little Feat :)

It was impromptu but well-received so maybe I should start doing this regularly at data science conferences…?

Video Tutorial on the Central Limit Theorem

Brand-new video tutorial: an intro to the Central Limit Theorem, a whacky phenomenon that underlies (almost) all modern statistics and machine learning. My hands-on code demos (in Python) make it easy to grasp why the whackiness happens.

Could be the perfect thing for ya if you're avoiding the plague by working from home and/or you're looking for something new to learn!

Announcing A4N: The Artificial Neural Network News Network →

It’s my great pleasure to announce A4N — the Artificial Neural Network News Network — a new podcast series. Our show is a lighthearted vehicle for discussion of the latest developments in A.I., machine learning, and data science, in which we both introduce technical aspects of these advances as well as their social implications.

The intended audience is anyone interested in automation, A.I., or the future, with brief sections catering especially to professionals working in the fields of data science or software engineering.

The show is hosted by Andrew Vlahutin, Grant Beyleveld, Vince Petaccio II, and myself — data scientists and machine learning engineers from untapt, an A.I. company. The podcast is commercial-free and we are bent on becoming your favorite — and most entertaining — source for A.I.-related news.

Episodes are hosted on jonkrohn.com (here) and are syndicated across all of the most popular platforms: Apple Podcasts, Spotify Podcasts, and Google Podcasts.

As with the Joe Rogan Experience and many other beloved podcasts, we also provide a raw video feed of the podcast recording session, which is available on YouTube.

Click through to the Medium post provided here for all of the links and more details :)

Tutorial, Booksigning, and "Ask the Expert" Session at ODSC East in Boston in April →

On April 15th in Boston, I’ll be giving a hands-on, 3.5-hour introduction to deep learning — featuring the brand-new TensorFlow 2 library — at the Open Data Science Conference East. In addition, I’ll be giving away dozens of free copies of my book Deep Learning Illustrated during a book-signing session, as well as answering any questions you might have during an “Ask the Expert” session.

First (of Many) Free Video Tutorials

After years of producing popular data-science tutorials that exist behind paywalls, I'm delighted to announce my first (of many) 100% free videos. This first one is on "Big O Notation" -- a critical computing concept.