Research Intern: Distributional Foundations of Multi-Agent Collaboration

Honda Research Institute USA; San Jose (CA); Sur place; USD 36 000 - 54 000; Plein temps

Résumé du poste

Honda Research Institute USA in San Jose, CA is offering a research internship focused on distributional foundations of multi-agent collaboration. You will develop mathematical models and algorithms to quantify coordination benefits in shared tasks, with emphasis on optimal transport and distribution-based agent behavior.

The role blends theory and computation: you will implement Python-based experiments, analyze results, and contribute to publications. Ideal candidates are pursuing a Ph.D.

Qualifications

• Currently pursuing a Ph.D. in a relevant field.
• Strong mathematical background including probability, linear algebra and optimization.
• Experience formulating and analyzing mathematical or computational models.

Responsabilités

• Develop and analyze mathematical models for quantifying the benefit of coordination in multi-agent systems.
• Apply optimal transport and related optimization methods to joint distributions over agent behaviours.
• Investigate how the benefits of coordination depend on task structure, agent behavior models & interaction geometry.
• Develop theoretical results and computational experiments characterizing properties of coordination metrics.
• Implement and evaluate algorithms in Python using numerical optimization and scientific computing tools.
• Design simulation experiments to validate theoretical predictions and extensions to human-robot settings.
• Contribute to research publications and presentations resulting from the work.

Connaissances

Mathematics

Probability

Linear algebra

Optimization

Python

Formation

Ph.D. in Computer Science, Robotics, Applied Mathematics, Electrical Engineering, Statistics, Operations Research or closely related field

Outils

Python

SciPy/NumPy

Description du poste

Research Intern: Distributional Foundations of Multi-Agent Collaboration - Honda Research Institute USA

Research Intern: Distributional Foundations of Multi-Agent Collaboration

Job Number: P26INT-12

Honda Research Institute USA (HRI-US) is seeking a self-motivated research intern to investigate the mathematical foundations of human-robot collaboration. The intern will develop methods for quantifying when and how much multi-agent systems can benefit from coordination, with a particular focus on optimal transport and distribution-based models of agent behavior. The work will combine mathematical analysis, algorithm development, and computational experiments to study fundamental properties of coordination in human-robot and multi-agent systems.

San Jose, CA

Key Responsibilities

• Develop and analyze mathematical models for quantifying the benefit of coordination in multi-agent systems.
• Apply optimal transport and related optimization methods to joint distributions over agent behaviors.
• Investigate how the benefits of coordination depend on task structure, agent behavior models, and interaction geometry.
• Develop theoretical results and computational experiments characterizing properties of coordination metrics.
• Implement and evaluate algorithms in Python using numerical optimization and scientific computing tools.
• Design simulation experiments to validate theoretical predictions and explore extensions to human-robot and multi-agent settings.
• Contribute to research publications and presentations resulting from the work.

Minimum Qualifications

• Currently pursuing a Ph.D. in Computer Science, Robotics, Applied Mathematics, Electrical Engineering, Statistics, Operations Research, or a closely related field.
• Strong mathematical background, including probability, linear algebra, and optimization.
• Experience formulating and analyzing mathematical or computational models.
• Proficiency in Python and scientific computing.
• Ability to independently develop, implement, and evaluate research ideas.

Bonus Qualifications

• N/A

Years of Work Experience Required 0

Desired Start Date May 2027

Internship Duration 3 Months

Position Keywords

Optimal transport; human-robot collaboration; multi-agent systems; coordination; coordination benefit; probability distributions;

convex optimization; distributional optimization; information theory; mathematical modeling; decision-making; robotics;

machine learning; statistical dependence