News
Correlation Doesn't Mean Causation! But What Does It Mean?
9+ hour, 12+ min ago (1041+ words) What does correlation tells us? Even before any of us got into data science, there was a phrase that we'd all heard; everyone knows it, young and old: It is a catchy phrase, and you've definitely said it once or…...
A Career in Data Is Not Always a Straight Line, and That's Okay
1+ day, 7+ hour ago (857+ words) Sabrine Bendimerad on why flexibility is a crucial data science skill, the risks of outsourcing human thinking to AI agents, and the changing terrain of career paths today. In the Author Spotlight series, TDS Editors chat with members of our…...
Your Synthetic Data Passed Every Test and Still Broke Your Model
5+ day, 10+ hour ago (957+ words) The silent gaps in synthetic data that only show up when your model is already in production. However, three months later, the fraud-detection model was failing to detect classes of transactions it had previously detected without fail, not just degrading…...
Correlation vs. Causation: Measuring True Impact with Propensity Score Matching
6+ day, 7+ hour ago (1197+ words) A step-by-step guide to reducing selection bias in Python Comparing groups is a common task in Data Science, especially if we are performing an A/B Test to understand the effects of a given variable over those groups. The problem…...
Ivory Tower Notes: The Methodology
6+ day, 10+ hour ago (1029+ words) A short intro to scientific methodology to combat "prompt in, slop'out" This is post #2 in my Ivory Tower Notes series. In post #1, I wrote about the problem: how every data and AI project starts.' This time, the topic is the…...
The LLM Gamble
1+ week, 1+ day ago (370+ words) Why it tickles your brain to use an LLM, and what that means for the AI industry When you open up the chat window for an LLM, and you have a question in mind, there's an undeniable sense of possibility....
Why MLOps Retraining Schedules Fail " Models Don't Forget, They Get Shocked
2+ week, 4+ day ago (1666+ words) Why calendar-based retraining fails in production, and how a practical shock-detection approach can work in real systems. Most production ML models don't decay smoothly " they fail in sudden, unpredictable shocks. When we fit an exponential "forgetting curve" to 555, 000 production-like fraud…...
How Does AI Learn to See in 3 D and Understand Space?
2+ week, 4+ day ago (1718+ words) But ask it to walk into an actual room and tell you which object sits on which shelf, how far the table is from the wall, or where the ceiling ends and the window begins in physical space " and the…...
How Visual-Language-Action (VLA) Models Work
2+ week, 5+ day ago (1725+ words) The mathematical foundations of Vision-Language-Action (VLA) models for humanoid robots and more How do robots understand the difference between raisins, green peppers and a salt shaker? More importantly, how can they figure out how to fold a t-shirt? That's the…...
Why AI Is Training on Its Own Garbage (and How to Fix It)
2+ week, 6+ day ago (927+ words) Deep web data is the gold we can't touch, yet If you have been interested in AI for a while, you are probably an LLM/Agent/Chat user, but have you ever asked yourself how these tools will be trained…...