Browsing by Author "Gökçesu, Hakan"

Now showing 1 - 5 of 5

Open Access
Asymptotically optimal contextual bandit algorithm using hierarchical structures
(Institute of Electrical and Electronics Engineers, 2018) Neyshabouri, Mohammadreza Mohaghegh; Gökçesu, Kaan; Gökçesu, Hakan; Özkan, Hüseyin; Kozat, Süleyman Serdar
We propose an online algorithm for sequential learning in the contextual multiarmed bandit setting. Our approach is to partition the context space and, then, optimally combine all of the possible mappings between the partition regions and the set of bandit arms in a data-driven manner. We show that in our approach, the best mapping is able to approximate the best arm selection policy to any desired degree under mild Lipschitz conditions. Therefore, we design our algorithm based on the optimal adaptive combination and asymptotically achieve the performance of the best mapping as well as the best arm selection policy. This optimality is also guaranteed to hold even in adversarial environments since we do not rely on any statistical assumptions regarding the contexts or the loss of the bandit arms. Moreover, we design an efficient implementation for our algorithm using various hierarchical partitioning structures, such as lexicographical or arbitrary position splitting and binary trees (BTs) (and several other partitioning examples). For instance, in the case of BT partitioning, the computational complexity is only log-linear in the number of regions in the finest partition. In conclusion, we provide significant performance improvements by introducing upper bounds (with respect to the best arm selection policy) that are mathematically proven to vanish in the average loss per round sense at a faster rate compared to the state of the art. Our experimental work extensively covers various scenarios ranging from bandit settings to multiclass classification with real and synthetic data. In these experiments, we show that our algorithm is highly superior to the state-of-the-art techniques while maintaining the introduced mathematical guarantees and a computationally decent scalability. IEEE
Open Access
Graph-theoretical dynamic user pairing for downlink NOMA systems
(IEEE, 2021-08-03) Köse, A.; Koca, M.; Anarım, E.; Médard, M.; Gökçesu, Hakan
We propose a novel graph-theoretical dynamic user pairing strategy based on the user rate requirements in cellular networks employing non-orthogonal multiple access (NOMA). The proposed approach relies on first constructing a conflict graph corresponding to all possible user pairings and then reformulating the problem of finding the best user pairs as that of finding the maximum weighted independent set (MWIS) on the conflict graph. This formulation turns the originally NP-hard problem into one that can be solvable in polynomial time thanks to the claw-freeness property of the conflict graph. The proposed user pairing method satisfies the maximum number of user demands with optimal network sum-rate as shown theoretically and as validated by the simulation results.
Open Access
Minimax optimal algorithms for adversarial bandit problem with multiple plays
(IEEE, 2019) Vural, Nuri Mert; Gökçesu, Hakan; Gökçesu, K.; Kozat, Süleyman Serdar
We investigate the adversarial bandit problem with multiple plays under semi-bandit feedback. We introduce a highly efficient algorithm that asymptotically achieves the performance of the best switching m-arm strategy with minimax optimal regret bounds. To construct our algorithm, we introduce a new expert advice algorithm for the multiple-play setting. By using our expert advice algorithm, we additionally improve the best-known high-probability bound for the multi-play setting by O(√(m)). Our results are guaranteed to hold in an individual sequence manner since we have no statistical assumption on the bandit arm gains. Through an extensive set of experiments involving synthetic and real data, we demonstrate significant performance gains achieved by the proposed algorithm with respect to the state-of-the-art algorithms.
Open Access
Qoe evaluation in adaptive streaming enhanced MDT with deep learning
(Springer, 2023-03-24) Gökçesu, Hakan; Erçetin, Ö; Kalem, G.; Ergut, S.
We propose an architecture for performing virtual drive tests for mobile network performance evaluation by facilitating radio signal strength data from user equipment. Our architecture comprises three main components: (i) pattern recognizer that learns a typical (nominal) behavior for application KPIs (key performance indicators); (ii) predictor that maps from network KPIs to application KPIs; (iii) anomaly detector that compares predicted application performance with said typical pattern. To simulate user-traces, we utilize a commercial state-of-the-art network optimization tool, which collects application and network KPIs at different geographical locations at various times of the day, to train an initial learning model. Although the collected data is related to an adaptive video streaming application, the proposed architecture is flexible, autonomous and can be used for other applications. We perform extensive numerical analysis to demonstrate key parameters impacting video quality prediction and anomaly detection. Playback time is shown to be the most important parameter affecting video quality, most likely due to video packet buffering during playback. We additionally observe that network KPIs, which characterize the cellular connection strength, improve QoE (quality of experience) estimation in anomalous cases diverging from the nominal. The efficacy of our approach is demonstrated with a mean-maximum F1-score of 77%.
Open Access
Sequential outlier detection based on incremental decision trees
(IEEE, 2019) Gökçesu, Kaan; Neyshabouri, Mohammadreza Mohaghegh; Gökçesu, Hakan; Serdar, Süleyman
We introduce an online outlier detection algorithm to detect outliers in a sequentially observed data stream. For this purpose, we use a two-stage filtering and hedging approach. In the first stage, we construct a multimodal probability density function to model the normal samples. In the second stage, given a new observation, we label it as an anomaly if the value of aforementioned density function is below a specified threshold at the newly observed point. In order to construct our multimodal density function, we use an incremental decision tree to construct a set of subspaces of the observation space. We train a single component density function of the exponential family using the observations, which fall inside each subspace represented on the tree. These single component density functions are then adaptively combined to produce our multimodal density function, which is shown to achieve the performance of the best convex combination of the density functions defined on the subspaces. As we observe more samples, our tree grows and produces more subspaces. As a result, our modeling power increases in time, while mitigating overfitting issues. In order to choose our threshold level to label the observations, we use an adaptive thresholding scheme. We show that our adaptive threshold level achieves the performance of the optimal prefixed threshold level, which knows the observation labels in hindsight. Our algorithm provides significant performance improvements over the state of the art in our wide set of experiments involving both synthetic as well as real data.