18 October 2019 (12:40):  IEEE AP/MTT/EMC/ED Turkey Seminar Series (S.54)

Speaker: Prof. Ekmel Özbay, Bilkent University

Topic: “A Review of Turkish GaN HEMT Technology Activities for RF Applications”

Location: Middle East Technical University, Ankara, Turkey

Abstract: In this talk, we will review the history, current status and future prospects of the Turkish GaN HEMT technology activities for RF and power electronics applications. After several years of academic research in this area, we have successfully launched a spin-off company and a joint venture between university and industry (Aselsan Bilkent Micro Nano Technologies Inc.) where we have developed the related GaN technologies that includes MOCVD growth, nanoscale fabrication, small and large signal circuit modeling, RF design, reliability and packaging for GaN based HEMT and MMICs that operate from 50 MHz to 40 GHz and beyond.

Bio: Prof. Dr. Ekmel Ozbay received the B.S. degree from Middle East Technical University and M.S. and Ph.D. degrees from Stanford University in electrical engineering, in 1987, 1989 and 1992. He worked as a postdoc in Stanford University and he worked as a scientist in Iowa State University. He joined Bilkent University (Ankara, Turkey) in 1995, where he is currently a full professor in Physics and EEE Departments. In 2003, he founded Bilkent University Nanotechnology Research Center (NANOTAM) where he leads a research group working on nanophotonics, nanometamaterials, nanoelectronics, and GaN based devices. He is the 1997 recipient of the Adolph Lomb Medal of  OSA and 2005 European Union Descartes Science award. He worked as an editor for Optics Letters, PNFA, SPIE JNP and IEEE JQE journals. He has published 495+ articles in SCI journals. His papers have received 17000+ SCI citations with an h-index of 58. He has given 165+ invited talks in international conferences. He recently became the CEO of a spin-off company: AB-MicroNano Inc.

vTools Link:

11 October 2019 (13:40):  IEEE AP/MTT/EMC/ED Turkey Seminar Series (S.53)

Speaker: Asst. Prof. Ercüment Çiçek, Bilkent University

Topic: “SPADIS: An Algorithm for Selecting Predictive and Diverse SNPs in Genome-wide Association Studies”

Location: Middle East Technical University, Ankara, Turkey

Abstract: Complex traits often cannot be explained by individual variants. Therefore, the efficient selection of multiple loci that explain the phenotype is critical for understanding the genetic basis of these traits. Selecting multiple loci is a computationally challenging problem that grows exponentially with the number of genomic variants. Many methods tackle this problem by focusing on coding regions to reduce the complexity of the problem. However, these approaches ignore the non-coding regions and introduce literature bias. As one alternative, regularized regression methods have been used; however, they do not allow the incorporation of background biological knowledge and suffer from long execution times. Currently, there is only one machine learning method in the literature, which aims to select a large set of loci efficiently by incorporating biological background information – SConES. SConES selects a set of features guided by a SNP-SNP network and favors the selection of SNPs that are connected on the network. We argue that while connectedness assumption is frequently used for functionally related features, it leads to the selection of redundant features when the goal is to explain a complex phenotype. In the current study, we hypothesize that selecting features on an SNP-SNP network that are diverse in term of location would correspond to incorporating complementary terms and thus, would help to explain the phenotype better. We present SPADIS that implements this novel idea by maximizing a submodular set function with a greedy algorithm that ensures a constant factor approximation to the optimal solution. We compare SPADIS to the state-of-the-art method SConES on a dataset of Arabidopsis Thaliana genotype and continuous flowering time phenotypes. We show that (i) SPADIS has better average phenotype prediction performance in 15 out of 17 phenotypes when the same number of SNPs are selected and provides consistent and statistically significant improvements in regression performance on average across multiple networks and settings, (ii) it identifies more candidate genes, and (iii) runs much faster compared to other methods. We also perform rigorous simulation experiments and compare SPADIS with off the shelf regression-based feature selection methods and show that SPADIS outperforms its counterparts.

Bio: Ercument Cicek earned his BS (2007) and MS (2009) degrees in Computer Science and Engineering from Sabanci University. He received his Ph.D. degree in Computer Science from Case Western Reserve University in 2013. During his Ph.D., he visited Cold Spring Harbor Laboratory to work on gene discovery algorithms for Autism Spectrum Disorder in 2012. After graduation, he worked as a Lane Fellow in Computational Biology at Carnegie Mellon University till 2015. Since then, he is an Asst. Prof. in the Computer Engineering Department of Bilkent University and is an adjunct faculty member in Computational Biology Department of Carnegie Mellon University. His research is mainly focused on designing machine learning algorithms for analysis of large-scale biological data. He is the recipient of Simons Foundation Autism Research Initiative (SFARI) Explorer Award, SFARI Pilot Award, TUBITAK Career Award, TUBA-GEBIP Award and Parlar Foundation Research Incentive Award.

vTools Link:

09-11 September 2019:  Technical Sponsorship for BEYOND 2019: Computational Science and Engineering Conference at Middle East Technical University, Ankara, Turkey

17 May 2019 (13:40):  IEEE AP/MTT/EMC/ED Turkey Seminar Series (S.52)

Speaker: Prof. F. Ömer Ilday, Bilkent University

Topic: “Ultrafast Laser-Driven Self-Organized Nano- and Micro-Structuring”

Location: Middle East Technical University, Ankara, Turkey

Abstract: Ultrafast laser processing has diverse applications, including creation of precision microstructures. However, scaling down to the nanoscale is strongly limited by the wavelength of the laser and by the cubic size of processing time with the spatial resolution, which we refer to, colloquially, as the “fat fingers problem,” the “explosion of complexity problem” and the increasingly strong thermal fluctuations.

The alternative approach that we are pursuing is to utilize laser-driven self-organization and self-assembly to put together the intended structures, which can be arbitrarily smaller than the wavelength of the laser beam. These structures are, then, dictated by the nonlinear dynamics of the system, which can be of a multitude of forms to be chosen from a limited, by rich library. We can choose them by adjusting only one or few parameters, typically, the laser power or polarization and different regions within a material can have different structures.

Our approach follows the principles laid out by luminaries like I. Progogine and H. Haken, already, in 1960s and 1970s, but was not applied to laser-material processing largely because much of the technologies we rely on did not yet exist. Our implementation is inspired by the physics of mode-locking of lasers, whereby modes that lock up in phase experience preferential “gain” over having random phases, which leads to a coherent structure in time. Similarly, we arrange for a certain coherent (typically periodic, but potentially aperiodic, as well) spatial structure to experience higher gain over the alternatives. In case of materials, this is achieved by driving the material locally far from thermodynamic equilibrium, which is necessary to gain access to multitude of spatial structures to choose from. Higher “gain” is achieved by invoking nonlinearities in the form of positive feedback between laser beam-induced changes in the material and material change-induced effects back on the laser beam.

We first showed that we could create laser-induced spatial nanostructures on various material surfaces with unprecedented uniformity (Ilday et al., Nature Photon., 2013). Afterwards, we have showed the benefits of nonlinear feedback in extremely efficient laser-material ablation (Ilday et al., Nature, 2016), creation of self-organized 3D structures inside silicon (Ilday et al., Nature Photon., 2017), and self-assembly of colloidal nanoparticles (Ilday et al., Nature Commun., 2017). We recently extended our results to self-assembly of colloids as small as a few nanometers, which is orders of magnitude smaller than the laser beam size that we use. We will also discuss how the symmetries of the feedback interactions determine the symmetries of self-organized patterns, and how we can use even “noise” to select desired patterns.

In these demonstrations, we have worked with physical systems that were deliberately chosen to be completely different from each case, from a silicon crystal to molybdenum surfaces or colloidal nanoparticles, to show that this approach is not material, size, or interaction specifics. Their commonality is that they are all nonlinear systems, which we deliberately drive far from equilibrium with the laser pulses. Although our focus is on understanding the basic physics, the talk will briefly showcase several applications.

Bio: Dr. F. Ömer Ilday received the BS degree in theoretical physics from Boğaziçi University, Istanbul, Turkey, in 1998. He took his PhD in applied physics from Cornell University, Ithaca, NY, USA, in 2003. He worked at Massachusetts Institute of Technology (MIT) from 2003 to 2006. In 2006, he joined Bilkent University as faculty member. He was awarded the European Research Council’s prestigious Consolidator Grant in 2013, the first consolidator grant and the first ERC grant on basic science in Turkey. Dr. Ilday graduated valedictorian of the top-ranked Physics Department at Bogazici University in 1998. In 2003, he received the prestigious RLE Fellowship from MIT. His contributions to science have been generously recognized through various awards, including Findlay Award from Cornell University (2004), Outstanding Young Scientist Award from the Turkish Academy of Sciences (TÜBA-GEBIP) (2006), Teşvik Award from the Scientific and Technological Research Council of Turkey (TÜBİTAK) (2011), Engin Arık Science Award from the Turkish Physical Society (2012) and the top award in science in Turkey, the Science Award of TÜBİTAK (2017). He is a full member of the Science Academy of Turkey and a senior member of the Optical Society (OSA).

vTools Link:

10 May 2019 (13:40):  IEEE AP/MTT/EMC/ED Turkey Seminar Series (S.51)

Speaker: Dr. Derya Malak, Massachusetts Institute of Technology and Northeastern University

Topic: “Coordinating Caching and Computation in Networks”

Location: Middle East Technical University, Ankara, Turkey

Abstract: This talk focuses on the central problem of coordinating computation and caching in networks, using some recent results in stochastic geometry and information theory. Our goal is to provide a FAST, RELIABLE, and CHEAP design for 5G mobile networks.  The first part of the talk focuses on decentralized caching by utilizing the redundancy across multiple, geo- dispersed, and mobile sources of data. In order to leverage proximity-based communications such as peer-to-peer systems or device-to-device communications, we exploited the spatial diversity of the content and the topology as a proxy for optimizing cache placement. We proposed novel decentralized and spatial exclusion-based cache placement policies. These policies promote diversity and reciprocation (FAST); provide guarantees on the cache hit probability (RELIABLE); and offload traffic from congested base stations, and are promising for proximity-based applications (CHEAP). The second part of the talk concerns with the limits of reliability with imperfect feedback when coding, and development of scalable and robust routing solutions for connectivity in wireless mesh networks. This approach utilizes coding for optimizing the tradeoff between in-order delivery delay and throughput, which is promising for computing systems such as the Internet of things, and ultra-reliable and low-latency communications e.g. mission-critical communications, and connected vehicles in 5G networks (FAST). It also provides robustness and delay guarantees (RELIABLE); and has very low complexity in terms of coding overhead, and is cost effective via the use of multi-hop WiFi links (CHEAP). Finally, this talk describes a new perspective on cloud/fog computing, by coordinating caching and computation in order to handle the large volume of data with growing computational demand. Our goal is to devise coding techniques for functional compression, and coordinating computation and caching in networks, by employing the concepts of graph entropy and function surjectivity. These techniques suit different applications such as caching, classification, federated learning, quantization, and compressed sensing. Our unified insights suggest to cache at the edge (FAST); distribute storage by exploiting geographic diversity and paths (RELIABLE); and distribute computation by making use of underlying redundancy both in data and functions, in order to recover a sparse representation, or labeling (CHEAP).

Bio: Derya Malak is a Postdoctoral Associate at the Massachusetts Institute of Technology and Northeastern University, where she has been working with Prof. Muriel Médard and Prof. Edmund Yeh, respectively. She received a Ph.D. in Electrical and Computer Engineering at the University of Texas at Austin under the supervision of Prof. Jeffrey G. Andrews, in August 2017, where she was affiliated with the Wireless Networking & Communications Group (WNCG). Previously, she received an M.S. degree in Electrical and Electronics Engineering at Koc University, Istanbul, Turkey, in February 2013. She received a B.S. in Electrical and Electronics Engineering (with minor in Physics) at Middle East Technical University, Ankara, Turkey, in June 2010. Derya has held summer internships at Huawei Technologies, Plano, TX, and Bell Laboratories, Murray Hill, NJ. She was awarded the Graduate School fellowship by the University of Texas at Austin between 2013-2017. She was selected to participate in the Rising Stars Workshop for women in EECS, MIT, in October 2018.

vTools Link:

03 May 2019 (13:40):  IEEE AP/MTT/EMC/ED Turkey Seminar Series (S.50)

Speaker: Assoc. Prof. Uluç Saranlı, Middle East Technical University

Topic: “Model-Based, Reactive Control of Legged Locomotion on Rough Terrain”

Location: Middle East Technical University, Ankara, Turkey

Abstract: Legged mobility has long been among key research areas in mobile robotics. In this context, accurate dynamic models of locomotory behaviors provide tools that are useful both in understanding biological systems as well as constructing robots and controllers to realize these behaviors. In this talk, I will focus on the latter, using spring-mass models that have been instrumental in the understanding and artificial realization of running behaviors. I will first describe our work in finding approximate analytic solutions for spring-mass models of running, followed by an application of these approximations in reactive footstep planning on rough terrain. Subsequently, I will describe a new, efficient method for energy regulation for such systems based on virtual tuning of leg damping that facilitates eventual physical realization of this model-based approach, characterizing performance through both simulations and experiments.

Bio: Dr. Uluç Saranlı is a Professor in the Department of Computer Engineering in Middle East Technical University, Ankara, Turkey. He received his B.S. degree in Electrical and Electronics Engineering from The Middle East Technical University, Turkey in 1996. He received his M.S. and Ph.D. degrees in Computer Science from The University of Michigan in 1998 and 2002, respectively. He then joined the Robotics Institute in Carnegie Mellon University as a postdoctoral associate until 2005. Before joining Middle East Technical University in 2012, he was an Assistant Professor in the Department of Computer Engineering in Bilkent University. His research interests focus on autonomous robotic mobility, with specific contributions in modeling, analysis, control of legged locomotion and behavioral planning for dynamically dexterous robot morphologies.

vTools Link:

26 April 2019 (13:40):  IEEE AP/MTT/EMC/ED Turkey Seminar Series (S.49)

Speaker: Dr. Ali Bayramoğlu


Location: Middle East Technical University, Ankara, Turkey

PDFs (in Turkish):




vTools Link:

12 April 2019 (13:40):  IEEE AP/MTT/EMC/ED Turkey Seminar Series (S.48)

Speaker: Assoc. Prof. Melda Yüksel, TOBB-ETÜ

Topic: “Precoder Design for Downlink Multiuser MIMO Systems”

Location: Middle East Technical University, Ankara, Turkey

Abstract: In this work, a downlink wireless communication channel is considered. The base station (BS) has common data for all users, unicast data for a set of intended users, and transmits the superposition of these messages. This setting neither falls into the non-orthogonal multiple access (NOMA) nor into the multi-group multicasting literatures. In NOMA systems, the BS has unicast data for all users, and multiple users share the same resources. In multi-group multicasting, there are non-overlapping groups, each demanding a different multicast message. This paper studies precoder design to achieve maximum weighted sum rate (WSR). It is first shown that the precoders designed for WSR maximization and weighted minimum mean square error (WMMSE) minimization are equivalent. Secondly, an iterative, low complexity algorithm (named as WMMSE), based on WMMSE transmit precoders and receivers, is proposed. Another low-complexity precoder, the phase aligned zero forcing (PAZF) precoder is also introduced. The results show that both algorithms converge fast. The WMMSE algorithm outperforms both PAZF and the zero-forcing (ZF) precoder for all signal-to-noise ratio (SNR) ranges. It offers better interference management and high coherent combining gains for common data, while PAZF finds the optimal phase rotation on the ZF precoder, and increases coherent combining gains.

Bio: Melda Yuksel received the B.S. degree in electrical and electronics engineering from Middle East Technical University, Ankara, Turkey, in 2001, and the Ph.D. degree in electrical engineering from Polytechnic Institute of New York University, Brooklyn, NY, in August 2007. She joined TOBB University of Economics and Technology, Ankara, Turkey, in Fall 2007, where she is currently an associate professor. Her research interests are in information theory, communication theory and wireless communications. Dr. Yuksel is the recipient of the best paper award at the Communication Theory Symposium of the 2007 IEEE International Conference on Communications and the Turkish National Science Foundation CAREER Award. Dr. Yuksel was the treasurer of 2013 IEEE International Symposium of Information Theory. She is currently serving as an editorial board member of PHYCOM, Physical Communication.

vTools Link:

05 April 2019 (13:40):  IEEE AP/MTT/EMC/ED Turkey Seminar Series (S.47)

Speaker: Prof. Abdullah Atalar, Bilkent University

Topic: “Research Ethics”

Location: Middle East Technical University, Ankara, Turkey

vTools Link:

27 March 2019: IEEE-MTT Distinguished Lecturer Seminar by Prof. Walid Ali-Ahmad

Topic: “Advanced RF FrontEnd and Transceiver Systems Design Overview for Carrier Aggregation based 4G/5G Radios”

Location: Sabancı University, İstanbul, Turkey

vTools Link:

22 March 2019: IEEE-MTT Distinguished Lecturer Seminar by Dr. Markus Gardill

Topic: “Automotive Radar – A Signal Processing Perspective on Current Technology and Future Systems”

Location: Middle East Technical University, Ankara, Turkey

vTools Link:

15 March 2019 (13:00):  IEEE AP/MTT/EMC/ED Turkey Seminar Series (S.46)

Speaker: Assoc. Prof. Selim Aksoy, Bilkent University

Topic: “Weakly Supervised Learning Algorithms for Medical Imaging and Remote Sensing Applications”

Location: Middle East Technical University, Ankara, Turkey

Abstract: Learning classifiers from large image sets has been a popular problem in computer vision and machine learning. The commonly employed supervised learning framework typically uses manually selected image patches with no ambiguity regarding their class labels. However, collecting sufficiently large number of examples for classes with high within-class variance and low between-class variance is not always possible. We will present weakly supervised learning algorithms for object recognition and image classification tasks in medical imaging and remote sensing applications with data sets having both localization and labeling uncertainties.

Bio: Dr. Selim Aksoy received the B.S. degree from Middle East Technical University in 1996, and the M.S. and Ph.D. degrees from the University of Washington, Seattle, USA, in 1998 and 2001, respectively. Since 2004, he has been with the Department of Computer Engineering, Bilkent University, where he is currently an Associate Professor. He spent 2013 as a Visiting Associate Professor at the Department of Computer Science & Engineering, University of Washington. His research interests include computer vision, pattern recognition, and machine learning with applications to remote sensing and medical imaging. He received the Research Incentive Award from the Prof. Dr. Mustafa Parlar Foundation in 2016, the BAGEP Young Scientist Award from the Science Academy Association in 2016, the GEBIP Outstanding Young Scientist Award from the Turkish Academy of Sciences in 2015, the Distinguished Teaching Award from Bilkent University in 2014, a Fulbright Scholarship in 2013, a Marie Curie Fellowship from the European Commission in 2005, and the CAREER Award from the Scientific and Technological Research Council of Turkey (TUBITAK) in 2004. He served as an Associate Editor of Pattern Recognition Letters during 2009-2013.

vTools Link:

08 March 2019 (13:40):  IEEE AP/MTT/EMC/ED Turkey Seminar Series (S.45)

Speaker: Assoc. Prof. Tolga Çukur, Bilkent University

Topic: “Rapid, Comprehensive, High-Resolution MR Imaging: From Sparse Recovery to Machine Learning”

Location: Middle East Technical University, Ankara, Turkey

Abstract: MRI offers an unprecedented opportunity to noninvasively examine the morphology and function of the human body in vivo. Yet, the quest for higher diagnostic utility by increasing image quality and diversity is often countered by limitations due to experimental and economic concerns. This talk will convey an overview of research at ICON Lab at Bilkent University towards addressing fundamental limitations to enable favorable trade-offs among imaging parameters. Technological innovations include high-resolution targeted pulse sequences, compressive sensing algorithms, as well as deep learning and other machine learning techniques for image processing and statistical modeling. These strategies can achieve substantial improvements in image quality for both structural and functional MRI. Challenging applications that involve the inverse problems of image reconstruction and image synthesis will be showcased.

Bio: Dr. Çukur received his B.S. degree from Bilkent University in 2003, and his Ph.D. degree from Stanford University in 2009, both in Electrical Engineering. He was a postdoctoral fellow at Helen Wills Neuroscience Institute at University of California, Berkeley till 2013. Currently, he is an Associate Professor in the Department of Electrical and Electronics Engineering, UMRAM, and Neuroscience Program at Bilkent University. His lab develops computational imaging methods for understanding the anatomy and function of biological systems in normal and disease states. He is the recipient of TUBITAK Career Award (2015), TUBA-GEBIP Outstanding Young Scientist Award (2015), BAGEP Young Scientist Award (2017), IEEE Turkey Research Encouragement Award (2017), Science Heroes Association Young Scientist of the Year Award (2018), and he is a senior member of IEEE (2017).

vTools Link:

01 March 2019 (13:40):  IEEE AP/MTT/EMC/ED Turkey Seminar Series (S.44)

Speaker: Asst. Prof. Emre Akbaş, Middle East Technical University

Topic: “Object Detection Through Search with a Foveated Visual System”

Location: Middle East Technical University, Ankara, Turkey

Abstract: In this talk, I will present a foveated object detector (FOD) as a biologically-inspired alternative to the sliding window (SW) approach which is the dominant method of search in computer vision object detection. Similar to the human visual system, the FOD has higher resolution at the fovea and lower resolution at the visual periphery. Consequently, more computational resources are allocated at the fovea and relatively fewer at the periphery. The FOD processes the entire scene, uses retino-specific object detection classifiers to guide eye movements, aligns its fovea with regions of interest in the input image and integrates observations across multiple fixations. Our approach combines object detectors from computer vision with a recent model of peripheral pooling regions found at the V1 layer of the human visual system. We assessed various eye movement strategies on the PASCAL VOC 2007 dataset and show that the FOD performs on par with the SW detector while bringing significant computational cost savings.

Bio: Dr. Emre Akbas is an assistant professor at the Department of Computer Engineering, Middle East Technical University (METU). Prior to joining METU, he was a postdoctoral research associate at the Department of Psychological and Brain Sciences, University of California Santa Barbara. He received his PhD degree from the Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign in 2011. His BS and MS degrees are from the Department of Computer Engineering, METU.

vTools Link:

15 February 2019: IEEE-APS Distinguished Lecturer Seminar by Prof. Ari Sihvola

Topic: “Metamaterials in electromagnetics: A bird’s-eye view”

Location: Middle East Technical University, Ankara, Turkey


vTools Link:

12 February 2019: IEEE-APS Distinguished Lecturer Seminar by Prof. Ari Sihvola

Topic: “Characterization and effective description of heterogeneous and composite electromagnetic materials”

Location: Gebze Technical University, Gebze, Turkey

vTools Link: