Understanding and Fostering Regional Artificial Intelligence Ecosystems: A Case Study in Maine

Harvard Data Science Review · 2023-10-27 · 2 citations

Artificial intelligence (AI) represents a novel force in both global and regional developments, transcending geographical, industrial, and academic borders.This article presents a case study in surveying AI challenges and opportunities in the state of Maine, in exploring ways to develop its AI ecosystem, and in fostering collaboration and development aligned with its strengths.It also showcases the potential when academics, government, and industry work together.

From Stochastic Parrots to Intelligent Assistants—The Secrets of Data and Human Interventions

IEEE Intelligent Systems · 2023-05-01 · 17 citations

Generative AI is all the rage nowadays—primarily driven by ChatGPT capturing the public imagination and attracting hundreds of millions of users in record time, reaching 100 million users in two months. However, there is much ambiguity from the providers about the technology, the methodology, and the way OpenAI makes it work. This compounds the mystique and speculation. I focus on what we know, with a particular emphasis on the aspects that the makers of ChatGPT avoid discussing with the public—namely, the underlying dependence on much manual intervention in training data curation, data labeling, operational interventions by humans, and reinforcement learning. Unfortunately, despite the criticality of these issues to the scientific community, they are hardly discussed. In this article, I attempt to address some of the issues in the hope of stimulating further studies of these less glorified but critical topics.

The Secrets of Data Science Deployments

IEEE Intelligent Systems · 2022-07-01 · 4 citations

Much attention is paid to data science and machine learning as an effective means for getting value out of data and as a means for dealing with the large amounts of data we are accumulating at companies and organizations. This has gained importance with the major waves of digitization we have seen, especially with the COVID-19 pandemic accelerating digital everything. However, in reality, most machine learning models, despite achieving good technical solutions to predictive problems wind up not being deployed. The reasons for this are many and have their origin in data scientists and machine learning practitioners not paying enough attention to issues of deployment in production. The issues range all the way from establishing trust by business stakeholders and users, to failure to explain why models work and when they do not, to failing to appreciate the importance of establishing a robust quality data pipeline, to ignoring many constraints that apply to deployed models, and finally to a lack of understanding the true cost of production deployment and the associated ROI. We discuss many of these problems and we provide what we believe is a pragmatic approach to getting data science models successfully deployed in working environments.

3rd IADSS Workshop on Data Science Standards - Hiring, Assessing and Upskilling Data Science Talent

Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining · 2022-08-12

The 3rd IADSS Workshop on Data Science Standards follows a tradition of two prior KDD workshops and the initial workshop at ICDM-2018. The theme of the 2022 workshop is: Hiring, Assessing and Upskilling Data Science Talent. Organized by the Initiative for Analytics and Data Science (IADSS.org) at KDD, the workshop provides a platform to discuss industry needs and practices around external and internal talent pipeline development in data science. We aim to provide an understanding of the data science job market, and the critical role of collaboration between academic institutions and industry to meet the increasing need for talent. IADSS conducts ongoing research in this domain. In this workshop, we share detailed findings and observations from this research. In addition to contribution from researchers and industry practitioners through an open call for papers, the workshop features several invited presentations and invited speakers. In order to achieve intended aim, interactive panels will discuss topics of interest and feedback from the workshop will be used to produce post-workshop learnings. This workshop is designed as a half-day working meeting with short talks, invited panels and discussion sessions to plan for future steps in the topic. Post conference, learnings from the workshop will be available at the workshop's home page: https://www.iadss.org/kdd2022

From Unicorn Data Scientist to Key Roles in Data Science: Standardizing Roles

Harvard Data Science Review · 2022-07-28 · 7 citations

The lack of an agreed-upon classification of job roles related to data science is causing much confusion that is challenging to the industry, educational sector, and practitioners. Prior work in this area has considered different aspects from different fields or points of view and has shown that more detail is needed in subcategorizing data science professionals. However, other prior work has also shown that avoiding the detailed subcategorization leads to challenging problems, for example, the pursuit of the elusive ‘data science unicorns.’ In this article, we target a simplification of prior work and an anchor categorization of job roles with clear definitions and expectations from each. We achieve this through analysis of survey results, LinkedIn profiles and job descriptions, and in-depth interviews with managing and hiring executives in data science. We also use our judgment as long-term practitioners and employers of data scientists to provide a practice-guided view of the problem.Our analysis has led to a simplification into three key role families with complementary skills: data analyst, data scientist, and data engineer. We believe this anchor categorization helps resolve several problems, including recruiting, forming, training, managing, and retaining effective data science teams. Although we realize there are and will continue to be many variations of these proposed anchor roles, this simplification is an effective tool to bypass the data science unicorn issue, and it can be used as a basis to establish more specialized or domain-specific roles. The combined skills in these role categories converge on the body of knowledge specification from the Initiative for Analytics and Data Science Standards (IADSS) data science knowledge framework (Fayyad & Hamutcu, 2020). The concise and familiar role categories simplify the problem and decompose it into more solvable subchallenges. We describe the essential knowledge required for each role and how, when, and in what ways it can be varied and extended. This description helps align expectations and serves as a step to tackle the pressing issue of training, evaluating, and building effective data science teams.

The Attention Economy and the Impact of Artificial Intelligence

Springer eBooks · 2021 · 22 citations

• Computer Science
• Sociology
• Business

Abstract The growing ubiquity of the Internet and the information overload created a new economy at the end of the twentieth century: the economy of attention. While difficult to size, we know that it exceeds proxies such as the global online advertising market which is now over $300 billion with a reach of 60% of the world population. A discussion of the attention economy naturally leads to the data economy and collecting data from large-scale interactions with consumers. We discuss the impact of AI in this setting, particularly of biased data, unfair algorithms, and a user-machine feedback loop tainted by digital manipulation and the cognitive biases of users. The impact includes loss of privacy, unfair digital markets, and many ethical implications that affect society as a whole. The goal is to outline that a new science for understanding, valuing, and responsibly navigating and benefiting from attention and data is much needed.

How Can We Train Data Scientists When We Can’t Agree on Who They Are?

Harvard Data Science Review · 2021-02-25 · 4 citations

How Can We Train Data Scientists When We Can't Agree on Who They Are? 2

Issue Information

Statistical Analysis and Data Mining The ASA Data Science Journal · 2020-11-02

Issue Information

Statistical Analysis and Data Mining The ASA Data Science Journal · 2020-03-10