About me

Mathematician and AI practitioner with 15+ years bridging rigorous research in real algebraic geometry, convex optimization, and deployed machine learning systems in government and industry. Author of the Irene optimization package and multiple open-source ML tools. Published researcher with 1,670+ citations (h-index 16) and deep expertise in translating mathematical theory into production AI solutions for public-sector challenges—including predictive analytics, resource optimization, and privacy-preserving deep learning.

Current Research

Mean Polynomial Cone Theory & Positivstellensatz — Generalizing SONC/SOS certificates of nonnegativity via weighted power mean forms for polynomial optimization.
Differential SDP (DSDP) — Extending Lasserre's moment/SDP hierarchy to optimization problems constrained by algebraic differential equations.
Privacy-Preserving Deep Learning — Homomorphic encryption methods for neural network inference over encrypted data (CryptoDL lineage).
Applied AI for Public-Sector Decision Making — Predictive analytics and generative AI systems for community safety and resource allocation.

Selected Scientific Libraries

Irene [15⭐]

General global optimization using algebraic methods. Implements a generalized Lasserre SDP hierarchy applicable to any optimization problem with a well-defined feasible region, solving truncated moment problems within commutative real algebras.

SKSurrogate

Automated ML pipeline optimization and hyperparameter tuning. Builds on scikit-learn's API for seamless integration into existing data science workflows.

Nonlinear Regression

Scikit-learn-style toolkit for nonlinear data relationships. Uses linear regression with data transformation and Hilbert space function spaces for weighted nonlinear regression.

A few other Python libraries on my GitHub:

pyProximation: Numerical Hilbert spaces, rational approximation, and collocation methods for systems of integrodifferential equations.
InventoryOptim: Forecasting, cost estimation, and trend optimization for dynamic multi-item inventory systems under limited, variable storage constraints.
Optimithon: An experimental implementation of some standard optimization methods.

Publications

Mathematics

R. E. Curto, M. Ghasemi, M. Infusino, S. Kuhlmann, The Truncated Moment Problem for Unital Commutative R-Algebras; Journal of Operator Theory 90 (2) (2023).
M. Ghasemi, M. Infusino, S. Kuhlmann, M. Marshall, Moment problem for symmetric algebras of locally convex spaces; Integr. Equ. Oper. Theory 90, 29 (2018). https://doi.org/10.1007/s00020-018-2453-7.
M. Alaghmandan, M. Ghasemi, Seminormed ⁎-subalgebras of ℓ∞(X); Journal of Mathematical Analysis and Applications 455 (1), 212-220 (2017).
M. Ghasemi, S. Kuhlmann, M. Marshall, Moment problem in infinitely many variables; Israel Journal of Mathematics 212, 1012-1012 (2016).
M. Ghasemi, M. Marshall, S. Wagner, Closure of the Cone of Sums of 2d-powers in Certain Weighted ℓ1-seminorm Topologies; Canadian Mathematical Bulletin 57 (2), 289-302 (2014).
M. Ghasemi, J. B. Lasserre, M. Marshall, Lower bounds on the global minimum of a polynomial; Computational Optimization and Applications 57, 387–402 (2014).
M. Ghasemi, S. Kuhlmann, M. Marshall, Application of Jacobiʼs representation theorem to locally multiplicatively convex topological R-algebras; Journal of Functional Analysis 266 (2), 1041-1049 (2014).
M. Ghasemi, M. Marshall, Lower Bounds for a Polynomial on a basic closed semialgebraic set using geometric programming; arXiv preprint arXiv:1311.3726 (2013).
M. Ghasemi, S. Kuhlmann, Closure of the cone of sums of 2d-powers in commutative real topological algebras; Journal of Functional Analysis 264 (1), 413-427 (2013).
M. Ghasemi, M. Marshall, Lower bounds for polynomials using geometric programming; SIAM Journal on Optimization 22 (2), 460-473 (2012).
M. Ghasemi, S. Kuhlmann, E. Samei, The moment problem for continuous positive semidefinite linear functionals; Arch. Math. 100, 43–53 (2013).
M. Ghasemi, M. Marshall, Lower bounds for a polynomial in terms of its coefficients; Archiv der Mathematik 95 (4), 343-353 (2010).
M. Ghasemi, M. Moniri, Beatty Sequences and the Arithmetical Hierarchy; Logic in Tehran: Proceedings of the Workshop and Conference on Logic, Algebra, and Arithmetic, Held October 18-22, 2003, Lecture Notes in Logic 26 (2006).

Computer Science

M. Ghasemi, D. Anvari, M. Atapour, J. S. Wormith, K. C. Stockdale, R. J. Spiteri The Application of Machine Learning to a General Risk–Need Assessment Instrument in the Prediction of Criminal Recidivism; Criminal Justice and Behavior 48 (4), 518-538 (2021).
E. Hesamifard, H. Takabi, M. Ghasemi, Deep Neural Networks Classification over Encrypted Data; CODASPY ’19: Proceedings of the Ninth ACM Conference on Data and Application Security and Privacy (2019).
E. Hesamifard, H. Takabi, M. Ghasemi, R. N, Wright, Privacy-preserving machine learning as a service; Proceedings on Privacy Enhancing Technologies (2018).
E. Hesamifard, H. Takabi, M. Ghasemi, CryptoDL: Deep Neural Networks over Encrypted Data; arXiv preprint arXiv:1711.05189 (2017).
E. Hesamifard, H. Takabi, M. Ghasemi, C. Jones, Privacy-preserving Machine Learning in Cloud; Proceedings of the 2017 on Cloud Computing Security Workshop, 39-43 (2017).
E. Hesamifard, H. Takabi, M. Ghasemi, Cryptodl: Deep neural networks over encrypted data; CoRR abs/1711.05189, arXiv preprint arXiv:1711.05189 (2017).
H. Takabi, E. Hesamifard, M. Ghasemi, Privacy preserving multi-party machine learning with homomorphic encryption; 29th Annual Conference on Neural Information Processing Systems (NIPS) 1, 4 (2016).
E. Hesamifard, H. Takabi, M. Ghasemi, CryptoDL: Towards Deep Learning over Encrypted Data; Annual Computer Security Applications Conference (ACSAC 2016), Los Angeles, California, USA (11), (2016).

Selected Experience:

Senior Scientist – Mathematician | Edmonton Police Service (2020–Present) Community Solutions Accelerator Lab.
Led AI/ML strategy for public-sector deployment, predictive analytics, and privacy-preserving ML pipelines for sensitive law enforcement data.
Optimization Modeling Consultant | Pfizer (2019–2020) Data Science Lab.
Developed mathematical optimization models for pharmaceutical supply chain and resource allocation challenges.
Data Scientist | Government of Saskatchewan (2017–2020)
Built ML-based risk assessment models for offender recidivism prediction deployed across correctional facilities. Developed forecasting tools for social services resource allocation.
Adjunct Professor | University of Saskatchewan (2018–Present)
Teaching advanced optimization, real algebraic geometry, and mathematical modeling. Mentoring graduate and undergraduate research.
MITACS Postdoctoral Fellow | University of Saskatchewan (2017–2019)
Joint project applying ML and optimization to criminology with the CS department, the Ministry of Justice, and the Saskatoon Police Service.

Education:

Ph.D. in Mathematics | University of Saskatchewan (2012). Research in Real Algebra, Moment Problem, and Optimization.

Postdoctoral Research Fellow | University of Saskatchewan (2014–2015). Optimization & convex algebraic geometry.

Postdoctoral Research Fellow | Nanyang Technological University, Singapore (2013–2014). Polynomial optimization, real algebraic/convex geometry, functional analysis.

Postdoctoral Research Fellow | University of Konstanz, Germany (2012). Real algebra, functional analysis, and polynomial optimization over invariant polynomials.

Page updated

Google Sites

Report abuse