Syrine Belakaria

Washington State University

Position: PhD Candidate
Rising Stars year of participation: 2021
Bio

Syrine Belakaria is a senior PhD student in Computer Science at Washington State University advised by Prof. Jana Doppa. Her research focuses on adaptive experiment design for science and engineering applications including hardware design materials design and electric transportation systems. She has been awarded the IBM PhD Fellowship (2021-2023) was a finalist for Microsoft Research Fellowship (2021) and Open Philanthropy AI Fellowship (2020) and won the Outstanding Research Assistant in EECS (2021) from the College of Engineering WSU. She has spent time as a research intern at Microsoft Research. Syrine is co-organizing the AAAI 2022 Workshop on AI to Accelerate Science and Engineering. She has published in top-tier AI venues including AAAI NeurIPS ICML and JAIR.

Output Space Entropy Search Framework for Multi-Objective Bayesian Optimization

Output Space Entropy Search Framework for Multi-Objective Bayesian Optimization
Many design optimization problems in science and engineering domains are instantiations of the following general problem: adaptive optimization of complex design spaces guided by expensive experiments where expense is measured in terms of resources consumed by the experiments. We consider the problem of black-box multi-objective optimization (MOO) using expensive function evaluations (also referred to as experiments) where the goal is to approximate the true Pareto set of solutions by minimizing the total resource cost of experiments. For example in hardware design optimization we need to find the designs that trade-off performance energy and area overhead using expensive computational simulations. The key challenge is to select the sequence of experiments to uncover high-quality solutions using minimal resources. In this work we propose a general framework for solving MOO problems based on the principle of output space entropy (OSE) search: select the experiment that maximizes the information gained per unit resource cost about the true Pareto front. We appropriately instantiate the principle of OSE search to derive efficient algorithms for the following four MOO problem settings: 1) The most basic single-fidelity setting where experiments are expensive and accurate; 2) Handling black-box constraints which cannot be evaluated without performing experiments; 3) The discrete multi-fidelity setting where experiments can vary in the amount of resources consumed and their evaluation accuracy; and 4) The continuous-fidelity setting where continuous function approximations result in a huge space of experiments. Experiments on diverse synthetic and real-world benchmarks show that our OSE search based algorithms improve over state-of-the-art methods in terms of both computational-efficiency and accuracy of MOO solutions.