Research

My research is at the interface of optimisation and machine learning. I seek solutions with a theoretically-provable performance.

Highlights

  • Query-Efficient Black-Box Adversarial Attacks: a framework for crafting adversarial examples based on gradient-sign estimation.
    100% evasion rate is achieved on MNIST with just 12 queries!
    [code, preprint]

  • Finite-Time Analysis for Lipschitz-Continuous Multi-Objective Problems [JOGO article]

  • BMOBench: A Black-box Multi-Objective Optimisation Benchmarking Platform [webpage].

  • NMSO: the Naive Multi-scale Search algorithm for expensive black-box optimisation.
    It was the second runner-up out of 28 algorithms in the BBComp’15 competition [INS paper, code].

  • MSO: a theoretical-analysis framework for multi-scale black-box search algorithms [JOGO paper].

Talks: