Postdoctoral Fellow: Decision‑Focused ML & Optimization

The Chronicle Of Higher Education, Inc.; Baltimore (MD); On-site; USD 63,480 - 100,000; Full time

Job summary

The Johns Hopkins University invites applications for a Postdoctoral Fellow to conduct research in decision-focused learning and optimization under uncertainty, with emphasis on end-to-end methodologies for robust scheduling and production planning. You will develop novel algorithms, perform computational experiments with real-world datasets, and contribute to publications and conference presentations.

A doctoral degree in a related field is required, with the opportunity to collaborate across

Qualifications

• Doctoral degree in a related field.
• Experience implementing computational experiments and using real-world datasets.
• Ability to publish in peer-reviewed journals and present at scientific conferences.

Responsibilities

• Conduct independent and collaborative research in decision-focused learning and optimization under uncertainty.
• Develop novel algorithms and computational methods that jointly learn predictive models and optimize decision-making.
• Investigate methods for incorporating uncertainty in optimization constraints, with applications in scheduling, production planning, and related operational systems.
• Implement and evaluate proposed methods using computational experiments and real-world datasets.
• Prepare manuscripts for publication in peer-reviewed journals and present research findings at scientific conferences.
• Collaborate with faculty investigators and research team members on related research initiatives.
• Participate in grant proposal preparation and other research activities as assigned.

Skills

Decision-focused learning

Optimization under uncertainty

Collaborative research

Education

Doctoral degree in a related field

Job description

The Johns Hopkins University invites applications for a Postdoctoral Fellow to conduct research in decision-focused learning and optimization under uncertainty, with emphasis on end-to-end methodologies for robust scheduling and production planning. You will develop novel algorithms, perform computational experiments with real-world datasets, and contribute to publications and conference presentations.

A doctoral degree in a related field is required, with the opportunity to collaborate across