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.

Experimentation 52

Experimentation Optimization Uncertainty Metrics 52

Smart Data Collective

FEBRUARY 4, 2020

Current R&D Models Provide Diminishing Returns. In a report on the failure rates of drug discovery efforts between 2013 and 2015, Richard K. Without better methodology, difficult-to-treat and ill-understood conditions and diseases are at risk of staying that way. To understand why, we must first understand the process.

Experimentation 100

Experimentation Testing Modeling Strategy 100

The Unofficial Google Data Science Blog

JULY 22, 2020

One reason to do ramp-up is to mitigate the risk of never before seen arms. A ramp-up strategy may mitigate the risk of upsetting the site’s loyal users who perhaps have strong preferences for the current statistics that are shown. For example, imagine a fantasy football site is considering displaying advanced player statistics.

Experimentation 95

Experimentation Statistics Testing Strategy 95

Occam's Razor

JANUARY 9, 2012

I've discovered that if we can just get them to imagine a better existence, undertake serious risks, experiment with new better ideas, and spend money executing them… they will ask for more robust measurement! Our mental model has not shifted enough to the existing reality. AND you can control for risk!

Marketing 140

Marketing B2B Measurement Risk 140

Domino Data Lab

JULY 2, 2019

Paco Nathan ‘s latest article covers program synthesis, AutoPandas, model-driven data queries, and more. Using ML models to search more effectively brought the search space down to 102—which can run on modest hardware. For details, see their SIGMOD 2015 paper where Michael Armbrust & co. Model-Driven Data Queries.

Metadata 105

Metadata Machine Learning Data Science Data-driven 105

Domino Data Lab

FEBRUARY 4, 2019

Spoiler alert: a research field called curiosity-driven learning is emerging at the nexis of experimental cognitive psychology and industry use cases for machine learning, particularly in gaming AI. The ability to measure results (risk-reducing evidence). Ensure a culture that supports a steady process of learning and experimentation.

Data Science 66

Data Science Experimentation Data-driven Machine Learning 66

O'Reilly on Data

DECEMBER 9, 2019

Recall from my previous blog post that all financial models are at the mercy of the Trinity of Errors , namely: errors in model specifications, errors in model parameter estimates, and errors resulting from the failure of a model to adapt to structural changes in its environment. For example, if a stock has a beta of 1.4

Statistics 27

Statistics Uncertainty Risk Marketing 27