Eric and Wendy Schmidt AI in Science Postdoctoral Fellowship Program
Call for Postdoctoral Applications
The Michigan Institute for Data Science (MIDAS) is seeking applications for a new AI in Science Postdoctoral Fellowship Program at the University of Michigan (the Schmidt AI in Science program). Funding for the program comes from the Eric and Wendy Schmidt Postdoctoral Fellowship, a program of Schmidt Futures. This program provides outstanding early-career researchers with intensive training and research experience as they ready themselves for independent research in academia and other sectors. We aim to enable the substantive use of AI for breakthroughs in science and engineering, and cultivate global science leaders. Specifically, we will:
1) Catalyze creative and transformative applications of AI with the potential to lead to major scientific breakthroughs carried out by the Fellows and their mentors
2) Enable a broader U-M research community to adopt AI in imagining, planning, executing, and supporting research applications across a range of science and engineering domains
3) Provide outstanding training to the Fellows
4) Collaborate with other sites to maximize the impact of the program
Priority Application Deadline: 11:59 pm EST, Jan. 15, 2023: All applications received by this deadline will be given full consideration. Late applications may be considered until all slots are filled
Offers will start to be sent out on Feb. 28, 2023
Accepted Fellows will have two weeks to decide whether to accept the offer.
The expected start date will be between Aug. 1 and Sept. 15, 2023, negotiable for special cases.
The program supports Fellows who apply AI methodology to address significant research questions in science and engineering. AI is defined broadly to include machine learning, robotics, Bayesian inference, and simulation. Science and engineering includes mathematical sciences, physical sciences, earth and environmental sciences, basic biological sciences, and engineering. All other disciplines, including biomedical, translational, medical, healthcare research, and social sciences, are out of the scope. See FAQ for additional explanations.
Please note that there is a sister program at MIDAS, the Michigan Data Science Fellows (DS Fellows) Program, which focuses on responsible data science and AI to improve their scientific and societal impact, understanding society through the use of new data types, and novel analytics for healthcare intervention. Potential applicants whose research is out of scope of the Schmidt AI in Science program may consider applying to the DS Fellows program.
The initial appointment is for one year, starting in September 2023, and is renewable for one more year contingent on satisfactory performance. The program will provide an annual salary of $76,000 plus benefits to each Fellow.
Each Fellow must have a Science Mentor and an AI Mentor established at the time of application. These mentors must be tenure-track faculty, research scientists, or research faculty at U-M. These and selected additional program faculty will form an individualized Fellowship Committee for each Fellow. All Fellows and their mentors will form a close-knit community through collaborative learning and training activities and research collaborations. Fellows will work on the University of Michigan Ann Arbor campus. They will have space at MIDAS and within the research groups of their mentors. See the list of potential mentors here.
This fellowship program builds a cohort of 10 new Fellows each year. The program will develop and provide extensive AI in science training to each annual cohort of Fellows, including AI bootcamps, workshops, hackathons, collaborative learning sessions, journal clubs, and a research training curriculum. Many formal courses at U-M will allow Fellows to audit as needed. Some program funds are available for additional AI training and workshops at U-M or elsewhere.
Fellows are expected to successfully implement their research and training plans and actively participate in collaborative learning and other program activities. They are expected to spend at least 50% of time in-person at MIDAS, at Weiser Hall on the Ann Arbor central campus, so that they can form a close-knit learning community. In addition, Fellows will be expected to build professional service experience through contributing ~10% of their effort to programmatic activities including training, coordinating research collaboration and building research tools.
Faculty Mentors are expected to devote sufficient time to provide research and career guidance to the Fellows, meet with them regularly, attend individualized Fellowship Committee meetings, and help Fellows seek opportunities for career advancement. They are also required to cover all the costs of the Fellows’ research, including research space, lab facilities, computers, computing resources, data access, travel to research conferences (at least two each year), and publication costs. Office desk space will be provided in MIDAS, but the faculty Mentors may wish to make desk space available in their research groups as well. The Science Mentor, by default, will be responsible to ensure the availability of all of the above; but the specific arrangements and shared responsibilities will be determined by the two Mentors of each Fellow.
Faculty Mentors are also expected to actively participate in a selected subset of the Schmidt AI in Science program activities in MIDAS, including bootcamps, workshops, collaborative learning sessions, developing training curriculum, among others.
The Schmidt AI in Science program will organize research and networking activities and conferences that will bring together Fellows and some mentors from across all of their postdoctoral training sites. The Fellows and mentors are expected to fulfill the Schmidt program’s participation expectations.
Eligibility: Applicants must be outstanding, intellectually curious researchers who will be ready to pursue independent research positions after two years in the program. They should have a doctorate in a related area before the expected start date of their fellowship. They should have never had a full-time faculty position. Collaboration experience and strong communication skills will be viewed favorably.
Individuals already at U-M are eligible to apply only if they propose research projects that are significantly different from their current projects AND if they will switch to two new mentors.
We recognize that strength comes through diversity and actively seek and welcome both candidates and mentors with diverse backgrounds, experiences, and identities.
Mentors: An applicant will need to first secure interest from two U-M faculty members who agree to be their Science mentor and AI mentor respectively. See the FAQ for how to find mentors.
- An applicant may reach out to any U-M faculty member(s) to inquire.
- If one faculty member agrees to be a mentor, they may work with the applicant together to secure a second mentor.
- If needed, the program team will help identify AI mentors. Please email email@example.com.
Applicants are encouraged to consult with the mentors extensively about the research statement and career goals before submitting the application.
The Application Package. The application should be one .pdf file with the following components, each prepared with legible formatting, including at least 1 inch margin on all sides:
- The candidate’s CV.
- Research Statement, up to 2 pages, font size 10 and above, that includes
- A compelling vision and a plausible approach to achieving that vision. Note that the program seeks to promote research that is transformative and creative, with the potential to spur significant advances in the application of AI in science domains. The statement should also describe the feasibility of the initial scope of the proposed research; and its potential to grow into a sustained research program for an independent research career.
- Qualifications that make the applicant particularly suitable for this program;
- Essential citations used in the research statement should be included in the 2 pages.
- Individualized Development Plan (IDP), up to 2 pages, font size 10 and above, that will include:
- The applicant’s career aspirations and how this Fellowship will benefit those;
- What AI expertise the applicant will need to acquire during the Fellowship, and an initial learning plan that may consist of a variety of formal and informal approaches;
- How will the Program (including training and participation in a cohort) benefit the applicant and how will the applicant contribute to the Program.
- Responsible AI Statement, up to ½ page, font size 10 and above. From the rigor of study design and the explainability of models, to the unbiased use of data and algorithms, and to data sharing and making a research project fully reproducible, responsible AI is critical to both science and society. Applicants should describe their approaches to and focus on responsible AI research. While such issues may be particularly salient with human subjects or societal impacts, all researchers need to take into consideration the effect of flawed data, brittle analyses, and irreproducible results.
- Diversity, Equity and Inclusion (DEI) Statement, up to ½ page, font size 10 and above. Describe how the candidate’s life experiences, past activities or anticipated activities will contribute to the diversity, equity and inclusion of the program and to the U-M AI in Science community.
- A Letter of Support from each proposed faculty mentor. This is not a reference letter. Instead, this is similar to a letter of support that is included in a training grant, delineating the applicant’s qualifications, the significance of the research project, the fit with this program, and a plan to mentor the applicant and to facilitate their career development. The mentors should also provide an explicit commitment to providing sufficient research resources and to participate in the overall Schmidt AI in Science program (see “Expectations” above).
- Names and contact information of two or three additional references. The reference letters should be emailed separately, from letter writers, to firstname.lastname@example.org by the deadline.
Submission: online submission only, through our application form.
- Fit with the program: whether the proposed research vision and the specific project align with the Schmidt AIM program scope.
- Candidate qualification: the candidate’s past experience and accomplishments in both AI and domain research, as expressed in their CV, letters of support and reference letters. Equally importantly, the candidate’s potential to become a research leader of the next generation: their willingness to take risks, creativity and originality, curiosity, collaborative spirit, entrepreneurship, and the ability to think big, as reflected in their past accomplishments and proposed plans.
- Research statement: the vision, the approach, and initial feasibility, the potential of making groundbreaking discoveries, the potential of becoming a sustained research portfolio after the Fellow leaves the program.
- Skills development: the candidate’s potential to acquire strong AI skills for the proposed research and long-term impact in the field; evidence that the candidate and mentors have carefully thought about the needs for skills development and are committed to it.
- Mentors: whether the selection of mentor(s) is appropriate for the research project, the training plan, and career development, as well as consistency among the mentor’s letters and the explicit commitment of the Mentors in participating in the broader program.
- Diversity: Of both fellows and mentors, along multiple dimensions, including demographics and disciplines.
- For University of Michigan internal applicants: whether the proposed research is significantly different from the current research and whether the proposed mentors are different from current ones.
The University of Michigan is committed to a policy of equal opportunity for all persons and does not discriminate on the basis of race, color, national origin, age, marital status, sex, sexual orientation, gender identity, gender expression, disability, religion, height, weight, or veteran status in employment, educational programs and activities, and admissions.
For questions, please contact email@example.com.
The University of Michigan leads the nation among research universities in research expenditures and is a national model of a complex, diverse, and comprehensive public research university. Over 100 U-M graduate programs are ranked in the top 10 in the U.S. by U.S. News and World Report, including its AI graduate education.
The Michigan Institute for Data Science (MIDAS) is a reputed center of excellence in data science and Artificial Intelligence (AI). Our mission is to strengthen University of Michigan’s (U-M) preeminence in data science and AI, and enable their transformative use in a wide range of research disciplines to achieve lasting societal impact. Currently, MIDAS focuses on five areas (“pillars”): 1) enhancing the scientific and societal impact of data science and AI through responsible research; 2) enabling the use of novel data types to measure and understand society; 3) transforming health interventions through the adoption of cutting-edge analytics; 4) enabling the adoption of AI methods for science and engineering research breakthroughs; and 5) boosting emerging research priorities. MIDAS supports the five pillars through research activities, training programs, community building and partnership development.
MIDAS supports research through a number of approaches:
- We develop guidance and resources, together with our faculty, for reproducible, responsible and ethical data science and AI.
- We lead strategic research thrust areas, such as the NSF funded Center for Data-Driven Drug Development and Treatment Assessment;
- We develop and manage research resources to enhance the U-M research capacity, including pilot funding, dataset access and industry collaborations.
- We convene faculty across disciplines to develop groundbreaking research ideas; help faculty secure major grants, especially collaborative grants; and support faculty’s education and outreach efforts that are integral parts of their grants.
- We organize a large number of research events, including themed research colloquia and the annual data science and AI Summit.
MIDAS training activities focus on enabling researchers to use novel data types and cutting-edge analytics to tackle significant research challenges.
- We provide workshops and bootcamps to introduce data science and AI methods to biomedical researchers, social scientists, environmental scientists, and others.
- Our Michigan Data Science Fellows program forms the core of a Data Science and AI postdoc community on campus.
- We also run a Graduate Data Science Certificate Program.
More than 460 MIDAS affiliate faculty members come from all schools and colleges at U-M Ann Arbor campus, and from U-M Dearborn and Flint campuses. This makes MIDAS one of the largest, and one of the most scientifically diverse, data science institutes at a US university. The faculty collaborate extensively, with expertise encompassing theoretical foundations and a wide range of Data Science and AI methodology, and its applications in almost all research areas at U-M. Our community also includes staff scientists and >1000 graduate and undergraduate students in Data Science and related degree programs, and in student-run data science and AI clubs.
MIDAS promotes a diverse and inclusive data science community by giving spotlights to women and underrepresented minorities as committee members and event speakers, and organizing training and social events specifically for these underrepresented groups. These events include the Women+ Data Science career and research series, and summer academies for underserved high school and undergraduate students.
MIDAS works with a large number of industry, academia, government and non-profit organizations to ensure that data science and AI research is enabled by real-world data and inspired by real-world challenges, and that research outcomes are translated into products, services and policies for positive social change. Our External Partnership program offers opportunities for joint research and talent recruitment for industry and public sector organizations. Our Data for Social Good program uses cutting-edge research to support data-informed decision making for our partners, including the City of Detroit, the Native American tribal nations in Michigan, and the US Environmental Protection Agency. MIDAS is a member of the NSF Midwest Big Data Hub and provides leadership in several focus areas. We also work closely with academic data science organizations to foster a broader and collaborative research community and to maximize synergistic impact. We organize the annual Future Leaders Summit that convenes outstanding students and postdocs from around the country to promote responsible data science and AI and foster the next generation of research leaders.
The following is a partial list of faculty members who are interested in being AI mentors. However, applicants and potential Science mentors should feel free to contact those who are not in the list as well.
(Faculty members can still sign up to be listed as AI mentors. By signing up, you confirm that you have sufficient AI expertise and mentoring experience to guide the postdocs in AI skills development).
- Aerospace Engineering: Bernstein, Duraisamy, Gorodetsky, Raman
- Architecture: Del Campo
- Computer Science & Engineering: Ackerman, Banovic, Chai, Chowdhury, Dereziński, Fouhey, Hu, Jagadish, Koutra, Mihalcea, Prakash, Provost, Scott, L. Wang, X. Wang, Wellman, Wiens
- Electrical & Computer Engineering: Balzano, Fessler, Hero, Nadakuditi, Qu, Scott, Subramanian, Ying
- Information: Brooks, Budak, Card, Collins-Thompson, Jacobs, Jurgens, Hemphill, Romero, Schoenebeck, Tomkins
- Mathematics: Bloch, Forger, Veerapaneni
- Mechanical Engineering: Gavini, Huan
- Naval Architecture & Marine Engineering: Skinner
- Nuclear Engineering & Radiological Sciences: Y Z
- Physics: Evrard, Gull
- Robotics: Barton, Chestek, Gregg, Ozay, Tilbury
- Statistics: Gagnon-Bartsch, Chen, Ionidies, Levina, Nguyen, Regier, Shedden, Tewari, Y. Wang, Xu, Zhu