Project Positions
Deep Learning-Based Next-Generation Wireless Communication Systems
| Faculty Member | Ertugrul Basar |
| Project Name | Deep Learning-Based Next-Generation Wireless Communication Systems |
| Project Description | Assoc. Prof. Ertuğrul Başar’s project titled “Deep learning-based Next-generation Communication Systems” will focus on deep learning-based novel communication system designs and technologies for the 6th generation (6G) and beyond wireless communication networks |
| Project Start Date | 1.08.2022 |
| Project End Date | 1.08.2025 |
| Number of students to be funded | 4 PhD Stundets |
| Desired background & experience | Wireless Communications & Deep/Machine Learning & Physical-Layer Designs |
| From which program/s students can be admitted | Electrical and Electronics Engineering |
Interplay between organelle and chromosome segregation during cell division
| Faculty Member | Ayşe Koca Çaydaşı |
| Project Name | Interplay between organelle and chromosome segregation during cell division |
| Project Description | Eukaryotic cells divide through mitosis to produce new cells with identical DNA. For this replicated chromosomes have to be equally segregated to the daughter cell. In addition to segregation of chromosomes, segregation and/or reorganization of other organelles are crucial for fidelity of cell division. These events must be carefully coordinated and controlled for maintenance of cell-ploidy and viability. Our lab is particularly interested in understanding how mitosis communicates with segregation of other organelles. We are particularly focused at the cross-talk of lysosome segregation with the process of mitotic exit. The inherent asymmetry and genetic malleability of the unicellular budding yeast Saccharomyces cerevisiae, maintain it at the forefront of model systems in which to dissect cell cycle regulation. We use this excellent model organism in our research. Our research integrates cutting-edge microscopy, high-throughput genetics and state-of-the-art proteomics techniques to uncover complexities of mitosis. By revealing novel mechanisms of cell cycle control and genomic stability we hope to contribute discovery of novel targets for cell cycle related diseases such as cancer. |
| Project Start Date | 1.09.2022 |
| Project End Date | 1.09.2026 |
| Number of students to be funded | 3 |
| Desired background & experience | a strong interest in cell biology. |
| From which program/s students can be admitted | Molecular Biology and Genetics |
CardioStressCI: Prevention of Vascular Cognitive Impairment through Early Detection of Cardiovascular Diseases
| Faculty Member | Attila Gürsoy, Özlem Keskin |
| Project Name | CardioStressCI: Prevention of Vascular Cognitive Impairment through Early Detection of Cardiovascular Diseases |
| Project Description | Cardiovascular disease is a known risk factor for the development of cognitive impairment and dementia as we age, but the reasons for this connection at the level of genes and proteins are still unclear. The goal of CardioStressCI is to identify proteins linked to both disorders, using cellular stress as a guide, with the ultimate goal of informing new diagnostic and treatment methods. |
| Project Start Date | 5.1.2021 |
| Project End Date | 5.1.2024 |
| Number of students to be funded | 1 MSc and/or PhD |
| Desired background & experience | Strong programming skills, preferably familiar with bioinformatics, background in molecular biology (especially protein structure and dynamics) and interested computational biology. |
| From which program/s students can be admitted | Bio-Medical Sciences and Engineering, Computer Science and Engineering, Molecular Biology and Genetics, Computational Sciences and Engineering |
| Additional information | You will be using computational methods developed in COSBILAB to predict how proteins interact with each other, build protein-interaction networks, learn and use network biology algorithms/tools to relate proteins to diseases. |
Functional characterization of enhancer activity in late stage prostate cancer
| Faculty Member | Nathan Lack |
| Project Name | Functional characterization of enhancer activity in late stage prostate cancer |
| Project Description | This computational project will characterize how changes in enhancer activity drive late-stage prostate cancer. Working closely with international collaborator we will analyze and integrate large-scale functional genomic data. |
| Project Start Date | 1.09.2022 |
| Project End Date | 1.08.2025 |
| Number of students to be funded | There is funding for at least two PhD/MSc students |
| Desired background & experience | Ability to code in either R or Python; Strong understanding of molecular biology and functional genomic techniques; Ability to work independently; Strong problem solving abilities |
| From which program/s students can be admitted | Molecular Biology and Genetics, Bio-Medical Sciences and Engineering, Computational Sciences and Engineering, Computer Science and Engineering, Physics |
| Additional information | Please contact me at nlack@ku.edu.tr for additional information. |
ERC-BeyondMoore
| Faculty Member | Didem Unat |
| Project Name | ERC-BeyondMoore |
| Project Description | The ERC (European Research Council) project has the overall objective to solve software side of the Post-Moore’s Law problem. The project will develop parallel programming models, performance models, compiler and runtime technologies designed for hardware accelerators. The implementations will be tested on large scale applications coming from deep learning, graph analytics, and high performance computing. |
| Project Start Date | 2.08.2021 |
| Project End Date | 2.08.2026 |
| Number of students to be funded | 2-3 |
| Desired background & experience | Required profile for Postdoc Positions All applicants must hold a PhD degree in computer science or related fields. The applicants should also have a strong track record in scientific publication and dissemination. The applicant should be fluent in both spoken and written English. All applicants must have research expertise in at least two of the following subjects: Parallel programing or Accelerator programming (e.g., OpenMP, CUDA, MPI) Performance optimization and performance modeling Scalable graph neural networks and deep learning Computer architecture Operating systems or system software Sparse Computation and Sparse Matrices Required profile for PhD Positions BSc in Computer Science/Engineering or related fields Applicants should have strong knowledge in Operation Systems, Computer Architecture, Data Structures and Algorithms No MSc is required to apply for the PhD position |
| From which program/s students can be admitted | Computer Science and Engineering |
Blockchain Advances based on Game Theory and Cryptography
| Faculty Member | Alptekin Küpçü |
| Project Name | Blockchain Advances based on Game Theory and Cryptography |
| Project Description | MSC, PHD and POSTDOC positions are available. Up to 7500 TL monthly net salary plus other benefits (housing support, meal card, health insurance). |
| Project Start Date | 15.06.2020 |
| Project End Date | 15.06.2023 |
| Number of students to be funded | 3 |
| Desired background & experience | Computer Science and Engineering |
| From which program/s students can be admitted | Computer Science and Engineering |
Computational Modeling of Turbulent Spray Combustion of Alternative Liquid Fuels Under High Pressure Conditions
| Faculty Member | Metin Muradoglu |
| Project Name | Computational Modeling of Turbulent Spray Combustion of Alternative Liquid Fuels Under High Pressure Conditions |
| Project Description | In this project, we aim to investigate the characteristics of the alternative liquid fuels under the conditions relevant to the aerospace gas turbine combustion chambers. Additionally, we aim to draw the fundamental knowledge required to design robust and efficient aerospace engine operating with alternative liquid fuels. Aerospace gas turbine combustion chambers feature the operating conditions having high pressure ambient condition, highly swirling flows, and intense turbulentchemistry interactions. In aerospace gas turbine combustion chambers, the oscillating heat release rate and the pressure fluctuation may interact and results in combustion instabilities which may cause severe damages to the engine. Therefore, the effect of the liquid fuels on the performance of the aerospace gas turbines must be characterized and understood. In literature the effect of the fossil fuels has been studied extensively, however the studies on the effect of alternative liquid fuels under aerospace gas turbine operation conditions have been rather scarce. In this project, the effect of the alternative liquid fuels on the performance of the gas turbine combustion chamber, and the interaction of the alternative liquid fuels with the highly non-linear flow dynamics will be investigated using numeric models. An extensive parametric study will be performed on different operating conditions in order to draw the characteristics of the alternative liquid fuel under the conditions relevant to the aerospace gas turbine operating conditions. To do that, as a first step, high-fidelity numeric solution algorithms will be developed for the numerical solution of the turbulent spray combustion flames. Comprehensive validation studies will be performed for the assessment of the developed numerical methods. In this project two different turbulent combustion models will be employed: i) the transported probability density function (tPDF) model, and ii) the flamelet generated manifold method (FGM). In tPDF method, the highly nonlinear chemical source terms appear to be in closed form, and there is no intrinsic assumption on the structure of the turbulent flame. Thus, the tPDF method can accurately model turbulence-combustion interactions, and resulting local extinction and reignition, emission, soot formation etc. Additionally, tPDF allows to draw very detailed analyzes on the flame structure due to existence of a very large number of Lagrangian particles. However, the computational cost of the tPDF method is high compared to the other combustion models, thus its usage in industrial design applications is limited. FGM method relies on the assumption that multi-dimensional structure of the turbulent flames can be considered as the assemble of onedimensional laminar flames, which significantly reduces the computational time required for simulation of turbulent flames. In the literature FGM method has been applied to simulate a wide range of turbulent flames in variety of combustion regimes. The main motivation of using these methods is i) advanced features of the tPDF method will enable us to make a very comprehensive analyzes and to draw detailed characteristics of the alternative liquid fuels, which is difficult to achieve in the reduced combustion models if not impossible, ii) FGM method has the capability of predicting the main flame features (temperature, concentration of the main species) in reasonable computation cost, which makes it very attractive for industrial design applications. In this study the validity and performance of FGM method will be assessed for the alternative liquid fuel combustions. |
| Project Start Date | 15.11.2021 |
| Project End Date | 15.11.2024 |
| Number of students to be funded | 2 MS and 2 PhD |
| Desired background & experience | Strong background in thermofluids sciences and math is required, and some background in computational methods and CFD is a plus. |
| From which program/s students can be admitted | Mechanical Engineering, Computational Sciences and Engineering |
MiMoGraph: Microfluidic Molecular Communications with Graphene-based Two-dimensional Nanoscale Receivers for Internet of Nano Things
| Faculty Member | Murat Kuscu |
| Project Name | MiMoGraph: Microfluidic Molecular Communications with Graphene-based Two-dimensional Nanoscale Receivers for Internet of Nano Things |
| Project Description | The objective of MiMoGraph is to develop the first micro/nanoscale experimental test and validation platform for microfluidic molecular communication (MC) systems accompanied by a theoretical design, modeling and optimization framework, and to devise novel experimentally-validated low-complexity MC methods, such as modulation, coding, synchronization, and detection techniques. |
| Project Start Date | 1.06.2021 |
| Project End Date | 31.05.2024 |
| Number of students to be funded | 2 MSc and/or PhD |
| Desired background & experience | |
| From which program/s students can be admitted | Bio-Medical Sciences and Engineering, Electrical and Electronics Engineering, Materials Science and Engineering |
Fully convolutional networks for semantic segmentation using 3D fractal and Poincare maps
| Faculty Member | Çiğdem Gündüz Demir |
| Project Name | Fully convolutional networks for semantic segmentation using 3D fractal and Poincare maps |
| Project Description | This project proposes to employ fractal analysis to represent complex patterns of objects/regions in CT/MR scans and incorporates this fractal analysis into the design of a deep neural network. |
| Project Start Date | 1.12.2021 |
| Project End Date | 1.07.2023 |
| Number of students to be funded | 2 |
| Desired background & experience | Knowledge on ML and deep learning and experience in designing and training encoder-decoder networks is a plus. |
| From which program/s students can be admitted | Computer Science and Engineering, Electrical and Electronics Engineering, Computational Sciences and Engineering, Bio-Medical Sciences and Engineering |
SparCity
| Faculty Member | Didem Unat |
| Project Name | SparCity |
| Project Description | EuroHPC funded SparCity project aims at creating a supercomputing framework that will provide efficient algorithms and coherent tools specifically designed for maximising the performance and energy efficiency of sparse computations on emerging HPC systems, while also opening up new usage areas for sparse computations in data analytics and deep learning. More on project website: http://sparcity.eu/ |
| Project Start Date | 1.04.2021 |
| Project End Date | 29.03.2024 |
| Number of students to be funded | 2-3 |
| Desired background & experience | Required profile for Postdoc Positions All applicants must hold a PhD degree in computer science or related fields. The applicants should also have a strong track record in scientific publication and dissemination. The applicant should be fluent in both spoken and written English. All applicants must have research expertise in at least two of the following subjects: Parallel programing or Accelerator programming (e.g., OpenMP, CUDA, MPI) Performance optimization and performance modeling Scalable graph neural networks and deep learning Computer architecture Operating systems or system software Sparse Computation and Sparse Matrices Required profile for PhD Positions BSc in Computer Science/Engineering or related fields Applicants should have strong knowledge in Operation Systems, Computer Architecture, Data Structures and Algorithms No MSc is required to apply for the PhD position |
| From which program/s students can be admitted | Computer Science and Engineering |
Deep learning-enabled crowd density estimation for cell analysis in digital pathology and characterization of homologous recombination deficiency in high-grade ovarian serous carcinoma
| Faculty Member | Çiğdem Gündüz Demir |
| Project Name | Deep learning-enabled crowd density estimation for cell analysis in digital pathology and characterization of homologous recombination deficiency in high-grade ovarian serous carcinoma |
| Project Description | This project aims to automatically segment and characterize cell nuclei in pathology slides using deep learning models and discover correlation between the nucleus characteristics and the homologous recombination deficiency status in high-grade ovarian serous carcinoma. |
| Project Start Date | 1.11.2021 |
| Project End Date | 1.11.2024 |
| Number of students to be funded | 2 students |
| Desired background & experience | Experience in designing and training convolutional neural networks, including encoder-decoder networks, and working on cell segmentation is a plus. |
| From which program/s students can be admitted | Computational Sciences and Engineering, Computer Science and Engineering, Electrical and Electronics Engineering, Bio-Medical Sciences and Engineering |
Shape-preserving deep neural networks for instance segmentation in medical images
| Faculty Member | Çiğdem Gündüz Demir |
| Project Name | Shape-preserving deep neural networks for instance segmentation in medical images |
| Project Description | This project proposes new deep neural network designs that enforce the learning model to preserve object shapes in segmentation. |
| Project Start Date | 1.04.2021 |
| Project End Date | 1.04.2023 |
| Number of students to be funded | 1 student |
| Desired background & experience | Experience in designing and training convolutional neural networks, including encoder-decoder networks, is a plus. |
| From which program/s students can be admitted | Bio-Medical Sciences and Engineering, Computational Sciences and Engineering, Computer Science and Engineering, Electrical and Electronics Engineering |
2ND CHANCE (ERC Project)
| Faculty Member | Levent Beker |
| Project Name | 2ND CHANCE (ERC Project) |
| Project Description | medical device development |
| Project Start Date | 1.05.2022 |
| Project End Date | 1.04.2027 |
| Number of students to be funded | 2 MS, 3 PhD, 2 Postdoc |
| Desired background & experience | Electrical Engineering, Mechanical Engineering, Materials Science, Chemistry, Medicine, or related fields |
| From which program/s students can be admitted | Bio-Medical Sciences and Engineering, Chemistry, Electrical and Electronics Engineering, Materials Science and Engineering, Mechanical Engineering, Molecular Biology and Genetics |
| Additional information | contact PI with your CV for further details. web site: microdevices.ku.edu.tr |
Machine Learning and Extreme Value Theory based Ultra Reliable Communication for Wireless Control Systems
| Faculty Member | Sinem Coleri |
| Project Name | Machine Learning and Extreme Value Theory based Ultra Reliable Communication for Wireless Control Systems |
| Project Description | Safety critical control applications still cannot leverage the advances in Wireless Internet of Everything (IoE) for two main reasons: 1) Lack of methodologies for guaranteeing ultra-reliable low latency wireless communication based on a statistical model of the environment, 2) challenge of integrating ultra reliable communications into wireless control systems. In this project, our goal is to develop innovative methods based on machine learning and extreme value theory for wireless channel modeling, communication and control systems in the ultra-reliable region in 6G and beyond systems. Extreme value theory (EVT) is a unique statistical discipline that develops techniques and models to describe rare events. On the other hand, machine learning provides efficient solutions for the challenging problems that cannot be solved with traditional analytical approaches in the design of next generation communication systems of complex and heterogenous nature. Large amounts of data are used in the extraction of ultra-reliable channel statistics and the optimization problem in which ultra-reliable communication is integrated into the control systems is very challenging, thus, the use of EVT together with machine learning techniques may enable ultra-reliable control systems. |
| Project Start Date | 6.1.2022 |
| Project End Date | 6.1.2025 |
| Number of students to be funded | 4 Ph.D students, 1 postdoctoral researcher |
| Desired background & experience | Wireless networks, machine learning |
| From which program/s students can be admitted | Electrical and Electronics Engineering |
Active Photoelectrode Development for Green Hydrogen Production via Solar Overall Water Splitting
| Faculty Member | Sarp Kaya |
| Project Name | Active Photoelectrode Development for Green Hydrogen Production via Solar Overall Water Splitting |
| Project Description | Using solar light to split the water by photo-electrochemical means to obtain hydrogen, to store the hydrogen or to catalytically convert the materials that can be used as direct fuel, is a way to efficiently remove the energy problem. The main objective of the project is to develop a photoelectrochemical cell for solar hydrogen production via overall water splitting. |
| Project Start Date | 4.1.2022 |
| Project End Date | 7.31.2024 |
| Number of students to be funded | 1 MS and 1 PhD |
| Desired background & experience | Chemistry, Material Science, Material Physics, Chemical Engineering |
| From which program/s students can be admitted | Chemistry; Materials Science and Engineering; Physics |
SynergyNet: Internet of Energy with Blockchain, Smart Contract and Federated Learning
| Faculty Member | Öznur Özkasap |
| Project Name | SynergyNet: Internet of Energy with Blockchain, Smart Contract and Federated Learning |
| Project Description | Internet of Energy is an innovative and effective approach that integrates the concept of smart network and Internet technology. Unlike traditional centralized energy systems, distributed Energy Internet system with multiple components and communication requirements needs innovative technologies for reliability and efficiency. Emerging and promising, distributed blockchain, smart contracts, and distributed federated learning technologies offer new opportunities for decentralized Energy Internet systems. Our objective in the SynergyNet project is to develop effective system models, techniques and algorithms by applying innovative distributed blockchain, smart contract and distributed federated learning principles to key research problems and areas within the Energy Internet. SynergyNet project is funded by TÜBİTAK 2247-A National Research Leaders program research grant. Fully funded PhD student and Postdoctoral researcher positions are available. |
| Project Start Date | 10.1.2022 |
| Project End Date | 10.1.2025 |
| Number of students to be funded | 4 PhD Students, 1 Postdoctoral Researcher |
| Desired background & experience | Distributed systems, Internet of Energy, Blockchain, Intelligent systems, Distributed federated learning, peer-to-peer (P2P) networks. |
| From which program/s students can be admitted | Computer Science and Engineering |
| Additional information | For detailed information, please contact PI (oozkasap@ku.edu.tr) with your CV. |
Network design and management problems for social aid distribution operations
| Faculty Member | Barış Yıldız |
| Project Name | Network design and management problems for social aid distribution operations |
| Project Description | The lack of an infrastructure to facilitate the fast and economical distribution of social aid is one of the main obstacles to enhancing the scope and the quality of the social aid efforts, which constitute one of the main pillars of solidarity and unity in our societies. Especially in large cities, the difficulties related to last-mile delivery operations (high economic and environmental cost) results in spending a significant part of the aid resources on delivery expenditures or blocking a substantial in-kind donation potential. In this project, we aim to address this critical issue and develop a novel social aid distribution network that takes advantage of the emerging smart transportation applications. |
| Project Start Date | 4.1.2022 |
| Project End Date | 4.1.2025 |
| Number of students to be funded | 1 MS and 1 PhD students will be hired |
| Desired background & experience | Students are expected to have a solid background in math programming, algorithm design and basic data analysis. Candidates with coding experience and competence in data visualization tools will be given priority. |
| From which program/s students can be admitted | Industrial Engineering and Operations Management |