No items found.

Optimization

Finding optimal solution for AI tasks ↓
Optimization

Research on the intersection of mathematics, computer science, and electrical engineering with a clear mission: find the optimal solution using advanced computational methods.

Mathematical optimization is a field of study on the intersection of mathematics, computer science, and electrical engineering that deals with the selection of a best element out of a set with respect to some criterion. The elements of the set are known as feasible solutions and the criterion is known as the objective function. Over the past couple of centuries, much of the work in mathematical optimization has focussed on the case of a convex, time-invariant set of feasible solutions and convex, time-invariant objective functions. This special case has become the work horse of machine learning, artificial intelligence, and most fields of engineering.

Research Focus

In our basic research, we focus study extensions towards (1) certain smooth, non-convex feasible sets and objective functions and (2) time-varying feasible sets and objective functions. The smooth non-convex problems, known as commutative and non-commutative polynomial optimization, have extensive applications in power systems, control theory, and machine learning, among others. The same applications can often benefit from the time-varying extensions.

Particular examples of this include our papers at AAAI 2019 (https://arxiv.org/abs/1809.05870) and AAAI 2020 (https://arxiv.org/abs/1809.03550), which deal with time-varying optimization.

Our papers at AAAI 2021 (https://arxiv.org/abs/2006.07315) and in the Journal of AI Research (https://arxiv.org/abs/2209.05274), which deal with non-commutative polynomial optimization. Notably, we can get the present best results on the COMPAS dataset.

In a recent paper in Automatica (https://arxiv.org/abs/2110.03001), we are working on the control on non-linear systems under uncertainty.

AutoFair (Human-Compatible AI with Guarantees)

AutoFair (Human-Compatible AI with Guarantees)

Horizon Europe project for fair AI algorithms supported by the EU with 3,8 million Euros. Imperial College London, the Israeli Institute of Technology Technion and the National and Capodistrian University of Athens as well as partners from the industry collaborate on developing explainable and transparent algorithms.

Quantum Computing for Finance

Quantum Computing for Finance

There has been recently much interest in quantum computing, which holds the promise of speeding up numerous (but certainly not all) problems. We work with some of the world's largest financial institutions to explore the applications of quantum computing in the Financial Services Sector.

Optimization in Power Systems

Optimization in Power Systems

Recently, we have made major advances in optimization in power systems. This includes time-varying optimization with non-linear constraints of the alternating-current optimal power flows, as well as mixed-integer optimization in unit commitment.

Research Results

Predictability and Fairness in Social Sensing

Predictability and Fairness in Social Sensing

The manner in which agents contribute to a social-computing platform (e.g., social sensing, social media) is often governed by distributed algorithms. We explore the guarantees available in situations, where fairness among the agents contributing to the platform is needed.

Warm-starting Quantum Optimization

Warm-starting Quantum Optimization

Many optimization problems in binary decision variables are hard to solve. In this work, we demonstrate how to leverage decades of research in classical optimization algorithms to warm-start quantum optimization algorithms. This allows the quantum algorithm to inherit the performance guarantees from the classical algorithm used in the warm-start.

Hacking
human future

By clicking “Accept All Cookies”, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist in our marketing efforts.