Postdoctoral Fellow at the Secure and Fair Machine Learning Lab

The Digital, Data, and Design (D^3) Institute at Harvard is accepting applications for multiple postdoctoral fellows to work on research activities at our newly formed research labs. D^3 will have a formal launch in July 2022 with 10 newly minted labs that will provide cutting edge research faced by academics and practitioners. For more information on D^3, please visit https://d3.harvard.edu.

The postdoctoral fellows will work under the direct supervision of faculty Principal Investigators and Lab Manager of each lab. D^3 is looking for candidates with diverse backgrounds and/or new perspectives. There are no teaching requirements for these open positions.

The Secure and Fair Machine Learning (SAFR ML) Lab, led by HBS Professor Seth Neel and Harvard SEAS Professor Salil Vadhan, is seeking a Postdoctoral Fellow. The lab focuses on developing algorithms that allow data science practitioners to trade-off ethical considerations like privacy, interpretability, and bias with accuracy, and to mitigate the risks of overfitting. Recent works on fairness have included new definitions of statistical fairness that account for a more complex protected group structure or a more flexible notion of similarity, new algorithms for efficiently deleting user data from neural networks, the SOTA bounds for adaptive data analysis, and new techniques for differentially private optimization. Ensuring privacy and fairness in large-scale genomic analyses is a new research interest.

The successful candidates will work on projects that could include (i) private release of aggregate genomic data, (ii) data deletion & machine unlearning (iii) privacy risks in explainable models (iv) fair or interpretable machine learning. The successful candidates will leverage their strong theoretical and computational background and communication skills to engage in all stages of the research, including the design, theoretical analysis, implementation, evaluation, and demonstration on real-world datasets.

The ideal candidate will have:

  • Demonstrably strong research skills, ideally with publications in top venues in machine learning, artificial intelligence, or sister conferences (e.g., ICML, NeurIPS, ICLR, KDD, AAAI, IJCAI, UAI, FAccT, AIES, AIStat, ACMEC, WINE), and/or top-tier interdisciplinary journals (e.g., Nature family of journals, PNAS, Science).
  • Prior research experience related to privacy/security or algorithmic fairness

Basic Qualifications:

  • A Ph.D. or equivalent degree in computer science, statistics, mathematics, computational biology or other related quantitative domain. PLEASE NOTE: If you have obtained your Ph.D. in the past 12 months you must be able to provide a certificate of completion from the degree-granting institution OR a letter from the institute’s registrar stating all requirements for the degree have been successfully completed and should verify the date the degree has been conferred. No exceptions.
  • Proficiency in computer programming (Python)
  • Strong team player with excellent communication skills

Additional Qualifications:

  • Familiarity with genomics or medical informatics

Application Details:

Applications will be accepted until the position is filled. To apply please email the following d3@harvard.edu with the subject “SAFR ML Lab Postdoctoral Fellow”. Please do not contact lab faculty; if you have any questions, please contact d3@harvard.edu.

 All applications should include the following:

  • Curriculum vitae
  • Copy of academic records (unofficial records are acceptable)
  • 2-page research statement
  • Two research papers
  • Contact details of at least two references

Candidates may be asked to undergo an assessment as part of the interview process.

Additional Information:

This is a term position through June 30, 2023, with the strong possibility of renewal based on funding and performance. Relocation funding not provided.

The University requires all Harvard community members to be fully vaccinated against COVID-19 and remain up to date with COVID-19 vaccine boosters, as detailed in Harvard’s Vaccine & Booster Requirements. Individuals may claim exemption from the vaccine requirement for medical or religious reasons. More information regarding the University’s COVID vaccination requirement, exemptions, and verification of vaccination status may be found at the University’s “COVID-19 Vaccine Information” webpage: http://www.harvard.edu/coronavirus/covid-19-vaccine-information.

This role is offered as a hybrid (some combination of onsite and remote) where you are required to be onsite at our Boston, MA based campus. Specific days and schedule will be determined between you and your manager.

While we continue to monitor the evolving COVID-19 guidelines and restrictions, we appreciate your understanding and flexibility with our interview process. Please note that we will be conducting interviews virtually (phone and or Zoom) for selected candidates until further notice.

Culture of Inclusion: The work and well-being of HBS is profoundly strengthened by the diversity of our network and our differences in background, culture, national origin, religion, sexual orientation, and life experiences. Explore HBS work culture at https://www.hbs.edu/employment.

Commitment to Equity, Diversity, Inclusion, and Belonging

Harvard University views equity, diversity, inclusion, and belonging as the pathway to achieving inclusive excellence and fostering a campus culture where everyone can thrive. We strive to create a community that draws upon the widest possible pool of talent to unify excellence and diversity while fully embracing individuals from varied backgrounds, cultures, races, identities, life experiences, perspectives, beliefs, and values.

EEO Statement

We are an equal opportunity employer and all qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability status, protected veteran status, gender identity, sexual orientation, pregnancy and pregnancy-related conditions, or any other characteristic protected by law.