Level Up: AlexNet, DeepSeek R1, & Linear Algebra Badges! 🚀

I’m starting the weekend strong with 3 new badges on Deep-ML! This round was particularly diverse, covering everything from the foundations of math to the cutting edge of reasoning models.

🏅 New Certifications

💡 What I Learned

This batch of challenges really stretched my understanding in different directions:

📐 Linear Algebra: The Foundations

Revisiting Vector Spaces and diving deep into SVD (Singular Value Decomposition) and PCA (Principal Component Analysis) was refreshing. It’s easy to forget the pure math when you’re deep in PyTorch tensors, but proving these concepts from scratch really solidified my intuition for why they work.

🧠 DeepSeek R1: Reasoning & RL

This was the most intimidating part—the math behind Reinforcement Learning (RL) can be scary! But working through the DeepSeek R1 collection helped me demystify concepts like KL Divergence. It was satisfying to understand the mathematical guardrails that keep reasoning models on track.

🏛️ AlexNet: The Classic

Implementing the components of AlexNet was a great history lesson. It’s amazing to see how the choices made in that architecture paved the way for the modern ConvNets we use today.

My journey on Deep-ML continues!