Jon Krohn on Deep Learning Advancements, PyTorch Lightning, and Going Beyond ML
Deep learning is becoming commonplace, as more and more companies are looking to take a deeper dive into their data — and going beyond just machine learning. We recently spoke with Dr. Jon Krohn about his upcoming deep learning bootcamp, what tools he uses, what platforms he’s excited for, and some interesting use cases he’s seen.
Q: Last year, you did a bootcamp on machine learning — why jump to deep learning next? Is there a particular value in deep learning that you see?
There are lots of learning materials out there on machine learning in general. For me, there’s more value in learning (and in my case, creating!) content that delves into specific niches of ML that are both disproportionately impactful and disproportionately untapped.
The bootcamp I offered in 2021 on the four foundational subject areas (i.e., linear algebra, calculus, probability theory, and computer science) that provide a firm foundation for understanding and applying machine learning is an example of one such niche. Instead of studying ML in general, the curriculum touched on ML examples only to provide hands-on applications of the four foundational subject areas. By mastering these subject areas, one becomes an especially capable and valuable ML practitioner so they are worth investing time in understanding.
My deep learning bootcamp is similar. It again doesn’t cover ML in general, but rather it focuses on the deep learning subfield of ML in particular. With the abundance of cheap compute and data storage of recent years (and the even cheaper compute and storage of years to come), deep learning is uniquely positioned amongst ML subfields to make real-world breakthroughs across applications as diverse as machine vision, natural language processing, artistic creativity, and complex sequential decision-making. Again, an area of ML especially worth understanding!
Q: Would you consider deep learning to be an “advanced” data science skill, or is it approachable to newcomers/novice data scientists?
Following the pedagogical approach of my book Deep Learning Illustrated, my visual, intuitive, and hands-on approach to teaching deep learning makes it unusually approachable to novice data scientists. So, while deep learning is an advanced skill with state-of-the-art applicability, you don’t necessarily need to be a highly experienced data scientist to make outstanding practical use of it.
Q: What tools/frameworks dominated 2021?
After a few years of catching up, in 2021 PyTorch finally overtook the TensorFlow/Keras combination as the most popular library for architecting and training deep learning models. Check out my recent talk on the relative strengths and weaknesses of these libraries for a sense of which to use or learn about first depending on your application needs.
Q: What newer tools/frameworks are you looking forward to learning about or using in 2022?
PyTorch Lightning is enticing as a lightweight wrapper around PyTorch code for easily scaling up models to training on lots of data or to deploying large models into performant production systems.
Q: What’s a fun recent case study you can describe where you’ve used deep learning?
At my day job at Nebula, we use deep learning to “understand” natural language on resumes and in job descriptions in order to automate human resources workflows, thereby enabling talented people to land the right opportunities for them more rapidly than ever before. You can check out a webinar I gave on this particular application here.
More on the bootcamp: This course with Dr. Jon Krohn is an introduction to deep neural networks that brings high-level theory to life with working, interactive examples featuring TensorFlow 2, Keras, and PyTorch — all three of the principal Python libraries for deep learning. Essential theory will be covered in a manner that provides students with a complete intuitive understanding of deep learning’s underlying foundations.
