50 Years of Data Science

By David L. Donoho. 2017

730 citations recorded by the source corpus.

ABSTRACT More than 50 years ago, John Tukey called for a reformation of academic statistics. In “The Future of Data Analysis,” he pointed to the existence of an as-yet unrecognized science, whose subject of interest was learning from data, or “data analysis.” Ten to 20 years ago, John Chambers, Jeff Wu, Bill Cleveland, and Leo Breiman independently once again urged academic statistics to expand its boundaries beyond the classical domain of theoretical statistics; Chambers called for more emphasis on data preparation and presentation rather than statistical modeling; and Breiman called for emphasis on prediction rather than inference. Cleveland and Wu even suggested the catchy name “data science” for this envisioned field. A recent and growing phenomenon has been the emergence of “data science” programs at major universities, including UC Berkeley, NYU, MIT, and most prominently, the University of Michigan, which in September 2015 announced a $100M “Data Science Initiative” that aims to hire 35 new faculty. Teaching in these new programs has significant overlap in curricular subject matter with traditional statistics courses; yet many academic statisticians perceive the new programs as “cultural appropriation.” This article reviews some ingredients of the current “data science moment,” including recent commentary about data science in the popular media, and about how/whether data science is really different from statistics. The now-contemplated field of data science amounts to a superset of the fields of statistics and machine learning, which adds some technology for “scaling up” to “big data.” This chosen superset is motivated by commercial rather than intellectual developments. Choosing in this way is likely to miss out on the really important intellectual event of the next 50 years. Because all of science itself will soon become data that can be mined, the imminent revolution in data science is not about mere “scaling up,” but instead the emergence of scientific studies of data analysis science-wide. In the future, we will be able to predict how a proposal to change data analysis workflows would impact the validity of data analysis across all of science, even predicting the impacts field-by-field. Drawing on work by Tukey, Cleveland, Chambers, and Breiman, I present a vision of data science based on the activities of people who are “learning from data,” and I describe an academic field dedicated to improving that activity in an evidence-based manner. This new field is a better academic enlargement of statistics and machine learning than today’s data science initiatives, while being able to accommodate the same short-term goals. Based on a presentation at the Tukey Centennial Workshop, Princeton, NJ, September 18, 2015.

Subfields: Artificial Intelligence, Management Information Systems, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Management Information Systems

In Statistics and Probability

Citations in the corpus

AI and the Everything in the Whole Wide World Benchmark (2021, methodology): “…to introduce practically-oriented and tightly-scoped AI tasks — namely, automatic speech recognition (ASR) or machine translation (MT) — where the required validation is whether the benchmark accurately reﬂects the practical task being asked of the computer in its real-world context [Donoho, 2017].”
Habitat: A Platform for Embodied AI Research (2019, background): “The development of Habitat is a long-term effort to enable the formation of a common task framework [12] for research into embodied agents, thereby supporting systematic research progress in this area.”
On Evaluation of Embodied Navigation Agents (2018, background): “Simulation supports the application of the common task framework [8] to embodied behavior.”
Playing for Benchmarks (2017, methodology): “Progress in computer vision has been driven by the systematic application of the common task framework [13].”
Statistical Literacy for Active Citizenship: A Call for Data Science Education. (2017, background): “…stakeholders hold different views on the question of whether data science constitutes an academic field on its own separate from statistics or rather is a subdomain of statistics (Donoho, 2015), a commonly accepted notion by Cleveland (2001) sees the field as the science of learning from data.”
Assessing Generalization in Deep Reinforcement Learning (2018, background): “Such testbeds have proven to be effective catalysts of concerted community-wide progress in other ﬁelds (Donoho, 2015).”
On the Robustness of Information-Theoretic Privacy Measures and Mechanisms (2018, background): “INTRODUCTION PUBLISHING of individual-level data is increasingly pervasive [2]–[4].”
AutoLearn — Automated Feature Generation and Selection (2017, background): “‘Predictive Modeling’ prefers to discuss only accuracy of prediction made by different algorithms on various data sets, but the crucial methodology here, driving success is the predictive culture’s secret sauce of feature engineering [2].”
The growing ubiquity of algorithms in society: implications, impacts and innovations (2018, background): “There are, however, mathematical reasons why we would assume that a simple model would yield good predictions [64] (see also [65]).”
Learning to Reason with Data: How Did We Get Here and What Do We Know? (2019, background): “A recent paper by Donoho (2017) helps explain this potential contradiction.”
FLASH: Fast Bayesian Optimization for Data Analytic Pipelines (2016, background): “2939829 which are considered to be an essential element in the data science field [11] due to their key role in large-scale data science projects.”
Rearrangement: A Challenge for Embodied AI (2020, background): “Clear task deﬁnitions and evaluation metrics are essential in the common task framework, which is substantially responsible for progress in computer vision and natural language processing [22].”
Re-evaluating evaluation (2018, background): “The classic train-and-test paradigm on common datasets, which has served the community well [5], is reaching its limits.”