Profile

    Dr. Hojjat Rakhshani

    Senior AI Researcher @ Decathlon Digital, Amsterdam

    I am currently focused on developing compact and efficient language models for multi-agent systems, with a strong emphasis on AI explainability. Previously, I worked on conducting research in the fields of optimization and AutoML as part of my Ph.D. dissertation at the University of Upper Alsace, culminating in several high-impact publications in top-tier international journals and multiple awards.

    Education

    Ph.D. in Artificial Intelligence

    Université de Haute-Alsace 2017-2020

    Dissertation: "The interplay of machine learning and metaheuristics"

    M.Sc. in Intelligence Systems

    University of Sistan & Baluchestan 2013-2016

    Specialized in Optimization

    Work Experience

    Senior AI Researcher

    Decathlon Digital May 2021-Present

    Research Engineer

    Université de Haute-Alsace Jul 2020 - Apr 2021

    Projects

    Semantic Search for Enhanced E-Commerce Navigation

    I spearheaded the development of an advanced AI-powered search engine to enhance the company’s e-commerce platform. This system leveraged state-of-the-art generative AI techniques to propose a semantic search engine capable of understanding user queries in context and delivering highly relevant product recommendations, significantly improving customer satisfaction and engagement.

    Store Layout Optimization with Rich Semantic Embeddings

    We developed AI-driven solutions to optimize inventory assortment, directly contributing to €80 million in additional revenue while reducing stock costs in physical stores. This involved aligning inventory management with market trends and sales forecasts, enabling Decathlon to balance supply chain efficiency with customer demand effectively.

    Turnover Forecasting with DeepAR

    We built a comprehensive forecasting model using Amazon SageMaker DeepAR to predict store and product turnover accurately. This also involes analyzing the impact of Covid-19 on store performance using XGBoost regression, providing actionable insights that informed Decathlon’s strategic business decisions during a critical time.

    AutoML for Scientific Article Relationship Analysis

    In this project, we developed a pipeline to identify and classify relationships between scientific articles which achieved 90% accuracy, significantly enhancing the efficiency of literature reviews and research synthesis. The results were recognized in the IEEE WCCI 2020 proceedings, highlighting its potential to transform academic workflows.

    Neural Architecture Search for Time Series Prediction

    This research focused on improving deep residual networks for time series forecasting through neural architecture search (NAS). This approach achieved state-of-the-art accuracy, surpassing leading benchmarks like HIVE-COTE. Published in the IJCNN 2020 proceedings, this work demonstrated how innovative neural architectures could provide more accurate and efficient time series analysis.

    Network Interdiction in Multi-Depot Routing

    We investigated network interdiction problems in multi-depot vehicle routing, developing optimization strategies to improve logistical efficiency even under disruptions. This research combined advanced mathematical modeling and combinatorial optimization techniques, offering practical solutions for real-world transportation and supply chain challenges.

    Optimization with Transfer and Ensemble Learning

    We designed a novel optimization framework that leveraged transfer and ensemble learning to reduce computational costs significantly. By retaining and reapplying knowledge from one problem to another, this approach improved efficiency without sacrificing performance, providing a powerful tool for tackling diverse optimization challenges.

    Crowd Movement Prediction with I3D Architecture

    We enhanced the Two-Stream Inflated 3D (I3D) deep learning architecture to predict crowd dynamics using the Crowd-11 dataset. By applying metaheuristic optimization, I fine-tuned the model for improved accuracy and real-time performance, contributing valuable insights to the field of crowd behavior analysis.

    Multi-Objective AutoML for Model and Algorithm Optimization

    We proposed a multi-objective optimization framework to automate machine learning model configuration. By balancing competing objectives such as accuracy, computational efficiency, and robustness, this framework streamlined the deployment of ML models tailored to specific applications, saving time and resources while maintaining performance.

    Awards & Honors

    Outstanding Dissertation Award

    University of Strasbourg

    2020

    First Prize for CG2019

    Oregon State University

    2019

    100% Ph.D. Scholarship

    French Ministry of Education

    2017

    Technical Skills

    Natural Language Modeling
    Synthetic Data Generation
    AI-Driven Personalization
    AutoML
    Retrieval Augmented Generation
    Cloud ML (AWS, DataBricks)
    Optimization
    MLOPs

    Participations

    Enterprise Tech Leadership Summit

    Amsterdam, Netherlands • 2024

    International Joint Conference on Neural Networks

    Glasgow, Schotland • 2020

    World Congress on Computational Intelligence

    Panel Moderator

    Rio de Janeiro, Brazil • 2020

    International Conference on Bioinformatics

    Madrid, Spain • 2018

    Selected Publications

    On the performance of deep learning for numerical optimization: an application to protein structure prediction

    Applied Soft Computing • 2021

    Neural architecture search for time series classification

    International Joint Conference on Neural Networks • 2020

    Automated machine learning for information retrieval in scientific articles

    IEEE Congress on Evolutionary Computation • 2020

    Speed up differential evolution for computationally expensive protein structure prediction problems

    Swarm and Evolutionary Computation • 2019

    Optimum design of tuned mass dampers using multi-objective cuckoo search for buildings under seismic excitations

    Alexandria engineering journal • 2018

    Snap-drift cuckoo search: A novel cuckoo search optimization algorithm

    Applied Soft Computing • 2017

    Hierarchy cuckoo search algorithm for parameter estimation in biological systems

    Chemometrics and Intelligent Laboratory Systems • 2016