Technical Blog
Deep dives into AI, ML, and software engineering concepts
Demystifying Direct Preference Optimization (DPO) with Simple Examples on LLMs
RLHF is powerful but complex. DPO simplifies alignment by eliminating the reward model — deriving the optimal policy directly from preference data using a clever mathematical reformulation.
From Sentiment to System Metrics: Measuring Societal Readiness with NLP
Large-scale technologies often fail not because they are technically impossible, but because society is not ready for them. This article shows how NLP can transform public discourse into a measurable Societal Readiness Level (SRL) for economic modeling.
Game Theories in AI and Beyond: The Shapley Value Explained
How do you fairly credit each player in a team effort? Shapley values — rooted in cooperative game theory — answer this question and have become one of the most principled tools for model explainability in AI.
OLS vs Time Fixed Effects: Why Model Choice Matters in Energy Markets After Fukushima
When we analyze real-world systems like energy markets, it’s tempting to start with a simple regression and move on. Discover why time fixed effects are often the minimum standard for rigorous analysis.