Michael Bidollahkhani

E-mailmichael.bkhani@uni-goettingen.de

Biography

Michael Bidollahkhani, under the supervision of Prof. Dr. Julian Martin Kunkel, is a dedicated computer engineer and machine intelligence researcher. His expertise in software engineering and automated systems is recognized by his Young Scientist award from the YSF of Iran in 2017 and 2023. As a member of the National Elites Foundation and the ACM, Michael is actively engaged in the development of advanced intelligent systems.
ORCID: 0000-0001-8122-4441
Google Scholar: https://scholar.google.com/citations?user=_rLezLYAAAAJ

Research Interests

  • Complex Systems
  • Computational Intelligence
  • Cognitive Modeling
  • Neural Information Processing
  • Neuroinformatics
  • Emergent Intelligence
  • Artificial Neural Networks
  • Cognitive Robotics

Projects

Advisory roles

  • 2024, Track chair, The Eighteenth International Conference on Advanced Engineering Computing and Applications in Sciences; AISys, ICSEA, CENTRIC tracks

Journal review duties

  • 2023, The American Journal of Artificial Intelligence (AJAI)
  • 2020, IEEE Signal Processing Society

Teaching

Summer Term 2025

Autumn Term 2024

Open Thesis Topics

Designing an Environmental Sustainability Labeling System for AI Services Based on Resource UsageApply

This thesis addresses the need for transparency in the energy (resource) consumption of AI services by proposing a standardized environmental sustainability labeling system. The goal is to analyze the energy consumption and computational load of different AI tasks (such as classification, generation, or scheduling) and translate them into simple, interpretable labels similar to those used in home appliances (e.g., A++ to E). The student will collect runtime and resource usage data for various AI models, evaluate their environmental impact, and propose a standardized method to display this information to users and developers.

Michael Bidollahkhani AI SustainabilityEnergy EfficiencyEnvironmental LabelingResponsible AI

Meta Machine Intelligence (MMI) for Error Detection in High-Performance Computing SystemsApply

Focusing on how context-sensitive AI models can improve the early detection of faults in high-performance computing (HPC) environments. The objective is to implement an adaptive mechanism that selects among pre-trained machine learning models based on system state, workload behavior, and error patterns. The research involves defining relevant system contexts, integrating multiple detection models, and evaluating their effectiveness in different runtime conditions using real or simulated log datasets. The outcome aims to increase fault detection reliability while maintaining scalability across heterogeneous HPC architectures.

Michael Bidollahkhani Error DetectionHigh-Performance ComputingAdaptive AISystem MonitoringMeta-Machine Intelligence (MMI)

Multi-Model Job Scheduling for Mixed Computing EnvironmentsApply

This thesis focuses on designing a context-aware job scheduler powered by mutiple models to operate across heterogeneous computing environments—including cloud, edge, and HPC systems. The proposed system dynamically selects the most appropriate scheduling model based on job characteristics and system availability. The study will involve the development of an AI-based scheduler that adapts to varying resource types, historical job outcomes, and performance metrics. The research contributes to enhancing scheduling efficiency in multi-layered computing systems where task diversity and infrastructure constraints coexist.

Michael Bidollahkhani Job SchedulingAI for ComputingHeterogeneous SystemsPerformance OptimizationMeta-Machine Intelligence (MMI)

Lightweight AI for Detecting Irregular Behavior in Device LogsApply

This thesis aims to develop a minimal and efficient AI tool for identifying unusual behavior in log files generated by small-scale devices or sensors. The student will design a lightweight anomaly detection system optimized for environments with limited memory and compute capacity, such as embedded systems. Context indicators (e.g., temperature, timestamp frequency, error patterns) will be integrated to improve detection relevance. The resulting system will be suitable for real-world applications like monitoring edge devices or low-power IoT nodes.

Michael Bidollahkhani Anomaly DetectionEmbedded SystemsDevice LogsLightweight AI

Interactive Dashboard for Monitoring AI Performance in System MaintenanceApply

This project involves the design and implementation of a web-based dashboard that visualizes the decision-making process and behavior of AI models configurable to use different available data sources and tool-sets in system maintenance tasks. The student will focus on presenting tool selection, confidence levels, warning predictions, and input variations over time in an interpretable and user-friendly format. The dashboard is intended to improve trust and transparency in predictive maintenance systems, particularly in dynamic environments where multiple data sources, models or strategies are deployed simultaneously.

Michael Bidollahkhani VisualizationAI MonitoringWeb DashboardSystem Maintenance

Theses

  • Implementation of a Liquid Neural Network Control System for Multi-Join Cyber Physical ARM, Michael Bidollahkhani (Master's Thesis), Advisors: Ferhat Atasoy, Abdellatef Hamdan, 2023-06, BibTeX
  • Extract and mining government services, especially USO and their impacts on the development of rural communities using data mining algorithms and artificial intelligence, Michael Bidollahkhani (Bachelor's Thesis), Advisors: I. Soleimani, A. Shahbahrami, 2016, BibTeX

Publications

2024

  • Poster: Predictive Maintenance in Server Farms with Time Series Analysis (Michael Bidollahkhani, Julian Kunkel), 2nd NHRConference 2024 at NHR4CES@TUDarmstadt, Darmstadt, Germany, 2024-09-09 BibTeX URL PDF
  • HOSHMAND: Accelerated AI-Driven Scheduler Emulating Conventional Task Distribution Techniques for Cloud Workloads (Michael Bidollahkhani, Aasish Kumar Sharma, Julian Kunkel), IEEE Computers, Software, and Applications Conference, pp. 1-8, IEEE, IEEE, COMPSAC 2024, 2024-07 BibTeX
  • Distracted AI: Integrating Neuroscience-Inspired Attention and Distraction Learning in ANN (Michael Bidollahkhani, M. Raahemi, P. Haskul), 2024 20th CSI International Symposium on Artificial Intelligence and Signal Processing, pp. 1-8, IEEE, IEEE, AISP \myDOI{10.1109/AISP61396.2024.10475279}, 2024-02 BibTeX DOI
  • Comparing Fault-tolerance in Kubernetes and Slurm in HPC Infrastructure (Mirac Aydin, Michael Bidollahkhani, Julian Kunkel), Proceedings of the 18th International Conference on Advanced Computing (ADVCOMP 2024), pp. 40-49, Venice, Italy, ISSN: 2308-4499. ISBN: 978-1-68558-184-8, 2024 BibTeX URL
  • Revolutionizing system reliability: The role of AI in predictive maintenance strategies (Michael Bidollahkhani, Julian Kunkel), IARIA CloudComputing 2024 Conference, pp. 1-9, Venice, Italy, ISSN: 2308-4294. ISBN: 978-1-68558-156-5, 2024 BibTeX URL

2023

  • LTC-SE: Expanding the Potential of Liquid Time-Constant Neural Networks for Scalable AI and Embedded Systems (Michael Bidollahkhani, Ferhat Atasoy, Hamdan Abdellatef), In arXiv preprint arXiv:2304.08691, 2023-04-18 BibTeX
  • A Novel Approach for Muscle Fatigue Disorders Detection Using EMG Based Time-Constant Neural Networks (Michael Bidollahkhani, F. Atasoy), In Gazi Journal of Engineering Sciences (2), 2023 BibTeX URL
  • LoRaline: A Critical Message Passing Line of Communication for Anomaly Mapping in IoV Systems (Michael Bidollahkhani, O. Dakkak, A. S. M. Alajeeli, B. S. Kim), In IEEE Access (11), pp. 18107-18120, 2023 BibTeX
  • Real-Time Building Management System Visual Anomaly Detection Using Heat Points Motion Analysis Machine Learning Algorithm (Michael Bidollahkhani, Isa Avci), In Tehnički vjesnik (30), pp. 318–323, 2023 BibTeX
  • Liquid Time-Constant Neural Networks (Michael Bidollahkhani), In Interdisciplinary Artificial Intelligence, Series: Nobel Scientific Works, Edition: 1, pp. 163 (Turkey), 2023 BibTeX URL

2018

  • The Neural Connection Spot (Michael Bidollahkhani, S. Darbarpanah), The International Conference on New Horizons in the Engineering Science, Istanbul, Turkey, 2018 BibTeX
  • The RPAT Algorithm for Politician Assessment and Evaluation (Michael Bidollahkhani, F. Bidollahkhani), The International Conference on New Horizons in the Engineering Science, Istanbul, Turkey, 2018 BibTeX

2017

  • Parallel programming Application on Medical Image Processing: MRI contours matching algorithm based on GPU accelerated methods for Tumor differential Analysis (Michael Bidollahkhani), International Congress on Science and Engineering, Hamburg, Germany, 2017 BibTeX
  • Subjects Extraction and text data classification by CHERNOFF algorithm and implementation with Java general-purpose computer programming language (Michael Bidollahkhani, M. F. Masouleh), IEEE Second National and First International Conference on Soft Computing, IEEE, IEEE, Guilan, Iran, 2017 BibTeX

2015

  • Optimization of Artificial Retina implant's vision (Michael Bidollahkhani, S. Darbarpanah), International Conference on Science and Engineering of Istanbul Technical University (ITU), Istanbul Technical University, Istanbul, Iran, 2015 BibTeX

2014

  • A Survey on Different Strategies on Preparing Data for Data Mining (Michael Bidollahkhani, M. F. Masouleh), ISC National conference on Soft computing at Guilan technical university, Guilan technical university, Guilan, Iran, 2014 BibTeX

All publications as BibTex