Welcome to Layer by Layer

Welcome to Layer by Layer

Hello and welcome to Layer by Layer - and experiment in learning in public where I’ll be sharing my exploration of machine learning concepts, techniques, and research.

Why I Started This Blog

The machine learning landscape is evolving at a breathtaking pace. From transformers to diffusion models, from reinforcement learning to neural radiance fields - there’s an overwhelming amount of innovation happening. As someone navigating this field, I often find myself wanting to document my learning journey, break down complex ideas, and create resources I wish had existed when I was starting out.

This blog is my attempt to:

1. Document my learning process - Writing helps me solidify my understanding
2. Develop and share practical tutorials - With code examples that actually work. This is inspired by a few of the ML blogging greats: Sebastian Raschka’s blog, Andrej Karpathy’s blog, Lilian Weng’s blog, and many more.
3. Summarize interesting papers - Distilling key ideas from research into accessible explanations (in the spirit of Two Minute Papers but in longer written form). This is inspired by Jay Alammar’s Illustrated Transformer his Illustrated Deep Seek R1 and overall his Illustrated* series now available at Substack.
4. Build in public - Showing both successes and the inevitable struggles, and also post project walkthroughs.
5. Link to resources - Occasionally, I’ll be posting links to useful resources - though I don’t want to make this another newslettter or news aggregator.

About Me

I’m Daniel Pickem, a researcher and engineer with a background in robotics and multi-agent systems. I completed my PhD at Georgia Tech where I developed the Robotarium, a remotely accessible swarm robotics testbed. My swarm robotics days are somewhat behind me - or maybe just on pause until swarm robotics has more practical applications. Until then, I am really excited and curious about machine learning - most of all foundation models of all modalities.

Robotarium

Join Me on This Journey

I hope you’ll find the content here useful, whether you’re just starting out in ML or are already well into your own journey. Feel free to reach out with questions, corrections, or suggestions for topics you’d like to see covered.

Let’s explore the fascinating world of machine learning together, one layer at a time.

Stay curious,

Daniel