Dariush conducts research broadly, but not limited to, in the following topics:
Machine Learning
In many real life applications, the volume of data to be handled has significantly increased due to the recent developments in information technologies.
Aside from the difficulties of storing such massive amounts of data, the processing of data usually has to be done in real-time. Hence, the need for machine learning algorithms
that can process the data on-the-fly with significantly low computational complexities is steadily growing. In this context, there exist significant practical and theoretical
difficulties to sequential learning, since there is usually no or little knowledge about the statistical properties of the underlying signals or systems involved.
Furthermore, the classical robust methods that are resilient to such challenges are overly conservative and usually static such that they provide profoundly inferior results
on the average, deeming them practically useless.
In order to provide robust adaptive methods that also perform satisfactorily in real life applications, we introduce sequential learning algorithms
that are mathematically guaranteed to work uniformly for all possible signals without any explicit or implicit statistical assumptions on the underlying signals or systems.
We construct adaptive algorithms that sequentially perform as well as the best batch algorithm, for any signal, that had the ability to observe the entire data in advance.
Unlike the state-of-the-art methods in the literature, our algorithms can tackle problems with non-stationarity in data modeling by sequentially learning the model structure.
In this manner, without introducing any ad-hoc assumptions or parameters, we introduce algorithms that directly minimize any desired loss measure in a strong deterministic sense.
Hence, our results are guaranteed to hold, not on the average, but in an individual sequence manner.
Publications on this topic include:
- D. Kari, F. Khan, S. Ciftci, and S. S. Kozat, "A Novel Family of
Boosted Online Regression Algorithms with Strong Theoretical Bounds," submitted to Machine Learning, Springer, 2016.
- N. D. Vanli, D. Kari, M. O. Sayin, and S. S. Kozat, "Predicting Nearly as well as The Optimal Twice Differentiable Regressor," Machine Learning, Springer, 2016. (under the fourth round of reviews)
- D. Kari, I. Marivani, I. Delibalta, and S. S. Kozat, "Boosted LMS-based Piecewise Linear Adaptive Filters," European Signal Processing Conference (EUSIPCO), Budapest, Hungary, 2016.
(IEEEXplore)
- D. Kari, and S. S. Kozat, "Enhancing The Performance of Recursive Least Squares Estimation Through Online Boosting," Accepted at the IEEE 13th International Colloquium on Signal Processing & Its Applications (CSPA), 2017.
- I. Marivani, D. Kari, and S. S. Kozat, "Online Anomaly Detection with Limited Feedback Using Accurate Distribution Learning," Accepted but not presented at IEEE Workshop on Machine Learning for Signal Processing (MLSP), 2016.
- D. Kari, I. Marivani, and S. S. Kozat, "Second Order Robust Adaptive Filters for System Identification in The Presence of Impulsive Noise," Accepted at the IEEE 13th International Colloquium on Signal Processing & Its Applications (CSPA), 2017.
Big Data Processing
Due to outstanding developments in sensor technologies, wide spread usage of smart phones and Internet, we now have the opportunity and capability to gather huge amounts of data in different real life signal processing applications (which was not possible in the past). Efficient and effective processing of this big data can significantly improve the performance of many signal processing algorithms. However, this big data has dimensions and volumes unseen before in signal processing problems, comes in different varieties and its quality, quantity and statistics rapidly change in time and among elements. To accommodate these problems, the big data should be adaptively processed by signal processing methods since the adaptive methods (1) can process the data online, i.e., instantly, without any storage requirement, (2) can constantly adapt to the changing statistics or quality of the data, hence can be robust and prone to variations and uncertainties.
In order to effectively process the big data, we aim to develop novel adaptive signal processing algorithms specifically tuned to effectively operate on large dimensional data and construct the big data adaptive signal processing framework, as the first time, by defining new cost measures and design methodologies.
Publications on this topic include:
- B. C. Civek, D. Kari, I. Delibalta, and S. S. Kozat, "Big Data Signal Processing Using Boosted RLS Alorithm," IEEE Conference on Signal Processing and Communications Applications (SIU), Turkey, 2016.
(IEEEXplore)
- F. Khan, D. Kari, I. A. Karatepe, and S. S. Kozat, "Universal Nonlinear Regression on High Dimensional Data Using Adaptive Hierarchical Tree," IEEE Transactions on Big Data, 2(2):175–188, June 2016.(IEEEXplore)
- D. Kari, M. M. Neyshabouri, and S. S. Kozat, "Efficient Implementation of Online Boosting Using
Bandit Setting". IEEE Signal Processing Letters, to be submitted, 2016.
- D. Kari, and S. S. Kozat, "Highly Efficient and Effective Hierarchical Online Regression Using Boosting", Knowledge and Information Systems, Springer, to be submitted, 2016.
Signal Processing for Communications
In many practical communications systems, the communications channel is usually unknown and possibly time-varying. The efficient communication over such systems requires
the estimation of the channel, where this estimate is usually subject to distortion. In this context, in order to provide reliable communications, robust estimation algorithms are needed.
However, the classical robust methods that are resilient to such distortions are overly conservative, hence, provide unacceptable results on the average.
In order to present robust algorithms without sacrificing from the average estimation performance, we propose novel robust approaches that minimize a worst case regret
that is defined as the difference between the estimation error and the smallest attainable estimation error with an LS or MMSE estimator. In this sense, we seek an estimator, whose performance is
as close as possible to that of the optimal estimator for any possible perturbation. By this formulation, we alleviate the highly conservative nature of the classical robust methods and introduce
robust algorithms that provide satisfactory performance on the average, unlike the classical approaches.
Publications on this topic include:
- D. Kari, N. D. Vanli, and S. S. Kozat, "Adaptive and Efficient Nonlinear Channel Equalization for Underwater Acoustic Communication,"
- D. Kari, I. Marivani, F. Khan, M. O. Sayin, and S. S. Kozat, "Robust Adaptive Algorithms For Underwater Acoustic Channels Estimation and Their
Performance Analysis," Digital Signal Processing, Elsevier, 2016. (under the second round of reviews)
- D. Kari, and S. S. Kozat, "A New Adaptive Low Complexity Turbo Equalization Algorithm Based on Adaptive Trees," IEEE Transactions on Communications, to be submitted, 2016.
|
