Saturday , June 23 2018
Home / Brad Delong, Berkeley / Must-See: Bin Yu: Three principles for data science: predictability, stability, and computability

Must-See: Bin Yu: Three principles for data science: predictability, stability, and computability

Summary:
Must-See: Why “three principles”? Why not “five principles”? Bin Yu: Three principles for data science: predictability, stability, and computability: “September 12, 2017 :: 4:10pm to 5:00pm :: 190 Doe Library” https://bids.berkeley.edu/events/three-principles-data-science-predictability-stability-and-computability Bin Yu: Three principles for data science: predictability, stability, and computability: “Prediction is a useful way to check with reality… http://delivery.acm.org/10.1145/3110000/3105808/p5-yu.pdf? …Good prediction implicitly assumes stability between past and future. Stability (relative to data and model perturbations) is also a minimum requirement for interpretability and reproducibility of data driven results (cf. Yu, 2013). It is closely related to uncertainty

Topics:
Bradford DeLong considers the following as important:

This could be interesting, too:

Delaney Crampton writes Weekend reading: “barriers to economic equality” edition

Equitable Growth writes Examining the links between rising wage inequality and the decline of unions

Bradford DeLong writes Brad DeLong: Worthy reads on equitable growth, May 25-31, 2018

Bridget Ansel writes A look at the motherhood wage gap on Mother’s Equal Pay Day

Must-See: Why “three principles”? Why not “five principles”?

Bin Yu: Three principles for data science: predictability, stability, and computability: “September 12, 2017 :: 4:10pm to 5:00pm :: 190 Doe Library” https://bids.berkeley.edu/events/three-principles-data-science-predictability-stability-and-computability

Bin Yu: Three principles for data science: predictability, stability, and computability: “Prediction is a useful way to check with reality… http://delivery.acm.org/10.1145/3110000/3105808/p5-yu.pdf?

…Good prediction implicitly assumes stability between past and future. Stability (relative to data and model perturbations) is also a minimum requirement for interpretability and reproducibility of data driven results (cf. Yu, 2013). It is closely related to uncertainty assessment. Obviously, both prediction and stability principles cannot be employed without feasible computational algorithms, hence the importance of computability

https://www.youtube.com/watch?v=xqBW8QKs9q4

Bradford DeLong
J. Bradford DeLong is Professor of Economics at the University of California at Berkeley and a research associate at the National Bureau of Economic Research. He was Deputy Assistant US Treasury Secretary during the Clinton Administration, where he was heavily involved in budget and trade negotiations. His role in designing the bailout of Mexico during the 1994 peso crisis placed him at the forefront of Latin America’s transformation into a region of open economies, and cemented his stature as a leading voice in economic-policy debates.

Leave a Reply

Your email address will not be published. Required fields are marked *