Â鶹¾«Ñ¡

• PhD in Computer Science, University Â鶹¾«Ñ¡, 2004
• BSc in Pure Mathematics, Sharif University of Technology (Tehran, Iran), 2000

Since obtaining his PhD in Computer Science in 2004 (University Â鶹¾«Ñ¡, UK) Amin has been working as a mathematician and computer scientist, in Sharif University of Technology, Tehran, Iran (2004-2006), Aston University, Birmingham, UK (2006-2009), University of Nottingham Ningbo China (2009-2015 and 2017-2023), and Halmstad University, Sweden (2015-2017). He is currently an Associate Professor in the School of Mathematics at the University Â鶹¾«Ñ¡.

Semester 2

LC Vectors, Geometry and Linear Algebra (Jinan)

Amin Farjudian supervises students in mathematical foundations of computer science.

Research Themes

• Domain Theory
• Robustness Analysis
• Ordinary and Partial Differential Equations
• Machine Learning

Research Activity

Amin Farjudian is interested in the interplay between mathematics and computer science, and the investigation of how the two disciplines enrich each other. He has published in several areas of mathematics and computer science, including domain theory, ordinary and partial differential equations, eigenvalue problems, symmetrisation, rearrangements of functions, and mathematical logic. His research is currently focused on developing domain-theoretic frameworks for analysis of mathematical problems arising in differential equations, machine learning, cyber-physical systems, and programming semantics.

The tight link between topology and computability is well-known and has been investigated extensively. Nonetheless, when the topological spaces involved do not have favourable properties (e.g., when they are not locally compact, or not core-compact) the classical methods are not adequate for laying a computational foundation. Such spaces arise in the study of some of the most important problems of classical and modern mathematics, e.g., functional analysis, ordinary and partial differential equations, probabilistic programming, and machine learning, to name a few. Amin's current research aims to address problems of this kind, through a combination of methods from domain theory, category theory, and classical real and functional analysis.

For instance, in a recent collaboration with Prof. Eugenio Moggi (University of Genoa, Italy) he has developed a domain-theoretic framework for robustness analysis of systems with state spaces that are not (locally) compact. Such systems include those that are modelled using differential equations, and also problems arising in kernel methods in machine learning.

In a related line of investigation, in collaboration with Professor Abbas Edalat (Imperial College London, UK) he has developed a domain model for solution of initial value problems (IVPs) with temporal discretisation. The topological spaces that arise in temporal discretisation of IVPs are not core-compact, hence the classical domain models are not adequate for addressing them. Thus, a novel approach had to be adopted for overcoming the challenge posed by temporal discretisation of IVPs, as appears in the common numerical methods such as Euler and Runge-Kutta methods for solving IVPs.

The long-term goal of Amin's research is establishing further links between the discrete world of Turing machines and the continuous world of mathematical analysis through the theory of domains.