2012 Experimentation Modeling 
The Unofficial Google Data Science Blog
APRIL 23, 2024
Experiments, Parameters and Models At Youtube, the relationships between system parameters and metrics often seem simple — straight-line models sometimes fit our data well. That is true generally, not just in these experiments — spreading measurements out is generally better, if the straight-line model is a priori correct.
Data Science and Beyond
JULY 22, 2018
I got my first data science job in 2012, the year Harvard Business Review announced data scientist to be the sexiest job of the 21st century. As I was wrapping up my PhD in 2012, I started thinking about my next steps. Things have changed considerably since 2012. What do I actually do here?
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
The Unofficial Google Data Science Blog
JUNE 30, 2016
In an ideal world, experimentation through randomization of the treatment assignment allows the identification and consistent estimation of causal effects. The choice of space $cal F$ (sometimes called the model ) and loss function $L$ explicitly defines the estimation problem. This is often referred to as the positivity assumption.
The Unofficial Google Data Science Blog
MAY 31, 2016
A geo experiment is an experiment where the experimental units are defined by geographic regions. The expected precision of our inferences can be computed by simulating possible experimental outcomes. The model regresses the outcomes $y_{1,i}$ on the incremental change in ad spend $delta_i$.
CIO Business Intelligence
MARCH 10, 2022
Fujitsu remains very much interested in the mainframe market, with a new model still on its roadmap for 2024, and a move under way to “shift its mainframes and UNIX servers to the cloud, gradually enhancing its existing business systems to optimize the experience for its end-users.”
O'Reilly on Data
JULY 25, 2023
I’m a professor who is interested in how we can use LLMs (Large Language Models) to teach programming. Here’s how I worked on it: I subscribed to ChatGPT Plus and used the GPT-4 model in ChatGPT (first the May 12, 2023 version, then the May 24 version) to help me with design and implementation.
The Unofficial Google Data Science Blog
OCTOBER 12, 2017
We data scientists now have access to tools that allow us to run a large numbers of experiments, and then to slice experimental populations by any combination of dimensions collected. Make experimentation cheap and understand the cost of bad decisions. This leads to the proliferation of post hoc hypotheses. Consider your loss function.
Expert insights. Personalized for you.
52 
Let's personalize your content