Deep Learning
-
The intuition behind neural networks and why they need activation functions.
20 min read -
Why Custom Inference in DeepStream?
10 min read -
Understanding how FPN allows deep learning models detecting small objects and how to implement it…
20 min read -
EmoNet: Speaker-Aware Transformers for Emotion Recognition — and What I’d Build Differently in 2026
Deep LearningA retrospective on my MS thesis, the leaderboard it placed on, and the LLM shift…
9 min read -
How did semantic search evolve from simple keyword matching into modern transformer-based language understanding? This…
23 min read -
Exploring the inner workings of a decoder-only Transformer foundation model
14 min read -
Solving multiplayer games with function approximation
9 min read -
A review of the Cross-Stage Partial Network paper — and a from-scratch PyTorch implementation
23 min read -
One scale parameter determines accuracy in rotation-based vector quantization.
7 min read -
PyTorch NaNs Are Silent Killers — So I Built a 3ms Hook to Catch Them at the Exact Layer
Deep LearningNaNs don’t crash your training — they quietly destroy it. After losing hours to a…
11 min read