Job Description

We are looking for a Research Fellow under the National University of Singapore (NUS) to support a project investigating the fundamental theoretical and algorithmic performance limits of Symbolic Regression (SR). SR is an emerging area of artificial intelligence that discovers interpretable mathematical relationships directly from data. You will lead the algorithmic and computational aspects of this project, conducting large-scale searches and benchmarking studies to characterize the best achievable performance of SR methods. This work contributes to explainable AI (XAI), AI for Science, and scientific discovery by advancing our understanding of interpretable machine learning models and their practical applications in real-world domains such as healthcare and science.

Key Responsibilities:
•    ⁠Scale: Design, develop, and manage large-scale computational pipelines for exhaustive search and optimization of symbolic regression models.
•    ⁠Benchmark: Conduct rigorous benchmarking and hyperparameter optimization of state-of-the-art symbolic regression algorithms and compare their performance against empirically derived performance limits.
•    ⁠Research: Contribute to the development and empirical validation of theoretical insights related to symbolic regression, including generalization and performance characterization.
•    ⁠Deploy: Assist in applying symbolic regression methods to real-world healthcare and scientific datasets to develop interpretable predictive models and equations.
•    ⁠Open Science: Curate, maintain, and release large-scale benchmark datasets, repositories, and research artifacts to support the global symbolic regression research community.
•    ⁠Dissemination: Publish research findings in leading machine learning and artificial intelligence conferences and journals, and contribute to project reporting and collaborative research activities.

Qualifications

•    A PhD in Computer Science, Electrical Engineering, Artificial Intelligence, Applied Mathematics, Statistics, or a related field. 
•    Coding: Proficiency in Python (NumPy, Pandas, Scikit-learn). Knowledge of C++ or SymPy is a major plus. 
•    Machine Learning: Strong understanding of machine learning fundamentals, including generalization, optimization, model evaluation, and statistical learning concepts. Experience with symbolic regression, evolutionary computation, AutoML, or interpretable machine learning is highly preferred.
•    Systems: Experience running experiments on HPC clusters or cloud environments. 
•    ⁠Documentation: Ability to use LaTeX and scientific writing tools for technical reporting and publication.
•    Analytical: Comfortable with mathematical theory (bounds, stability) and data-driven discovery. 
•    ⁠Proactive: Able to bridge the gap between theoretical research and scalable software engineering.
•    ⁠Track Record: Strong publication record in leading machine learning, artificial intelligence, data science, or evolutionary computation venues (e.g., ICML, ICLR, NeurIPS, AAAI, KDD, GECCO, or equivalent).
•    Open to Fixed Term Contract