Adilet is a young committed mathematician born and raised in Kazakhstan who wishes to further his academic experience and intends to attribute his future’s work to mathematical research and teaching. Adilet received his bachelor’s degree in mathematics from Nazarbayev University, a master’s degree in Applied Mathematics from the University of Manchester and a PhD degree in the field of Optimization from the University of Oxford in 2021. To pursue his PhD, he was awarded a competitive scholarship from the Alan Turing Institute, the UK’s national institute for data science and artificial intelligence.

Research:

His doctoral research lies at the boundary between optimization and machine learning. While optimization techniques are widely used in machine learning, can machine learning techniques be used in optimization? Inspired by this question, his research investigates the effectiveness of a particular machine learning technique called ‘dimensionality reduction’ applied to a high-dimensional global optimization problem, a problem notorious for its scalability challenges. In his work, he demonstrates that these challenges can be significantly alleviated for a certain class of functions, in literature called functions with low effective dimensionality. These functions appear in applications such as neural networks, combinatorial optimization problems, climate modelling and complex engineering and physical simulations.

In his future research, he is considering exploring the effectiveness of random methods in optimization with possible applications in machine learning, engineering and climate modelling problems. In particular, he would like to develop a more scalable global optimization algorithm by combining random search techniques with local optimization techniques. Furthermore, he is keen to investigate the practicality of random methods in other related fields such as, for example, linear sketching in the matrix completion problem to improve scalability of the existing matrix completion methods, many of which rely on expensive singular value decomposition.

Teaching:

Throughout his teaching career, Adilet has engaged in different formats of teaching including peer tutoring, online teaching, one-to-one tutoring, teaching in small and big classrooms. His teaching experience includes TA and tutorship roles in Continuous Optimization, Integer Programming and Graph Theory taught in the Mathematical Institute, Oxford.

To improve his teaching skills, he took part in the Developing Learning and Teaching workshop certified by Staff and Educational Development Association (SEDA). He successfully completed the course and received a SEDA award (for more info visit https://www.seda.ac.uk/).

C. Cartis, E. Massart, and A. Otemissov. Applications of conic integral geometry in global optimization. 2021. Research completed, in preparation

C. Cartis, E. Massart, and A. Otemissov. Constrained global optimization of functions with low effective dimensionality using multiple random embeddings. ArXiv e-prints, page arXiv:2009.10446, 2020. Manuscript (41 pages) submitted for publication in Mathematical Programming. Got back from the first round of reviewing

C. Cartis and A. Otemissov. A dimensionality reduction technique for unconstrained global optimization of functions with low effective dimensionality. Information and Inference: A Journal of the IMA, 2021. iaab011

X. Fan, A. Otemissov, F. Sica, and A. Sidorenko. Multiple point compression on elliptic curves. Designs, Codes and Cryptography, 83(3):565–588, 2017

D. Wei, M. Fyrillas, A. Otemissov, and R. Bekishev. Optimal design of helical springs of power law materials. ArXiv e-prints, page arXiv:1610.09155, 2016