Computer Science

Permanent URI for this collectionhttps://hdl.handle.net/10679/43

Browse

Now showing 1 - 20 of 542

Metadata only
The 13th international automated negotiating agent competition challenges and results
(Springer, 2023) Aydoğan, Reyhan; Baarslag, T.; Fujita, K.; Hoos, H. H.; Jonker, C. M.; Mohammad, Y.; Renting, B. M.; Computer Science; AYDOĞAN, Reyhan
An international competition for negotiating agents has been organized for years to facilitate research in agent-based negotiation and to encourage the design of negotiating agents that can operate in various scenarios. The 13th International Automated Negotiating Agents Competition (ANAC 2022) was held in conjunction with IJCAI2022. In ANAC2022, we had two leagues: Automated Negotiation League (ANL) and Supply Chain Management League (SCML). For the ANL, the participants designed a negotiation agent that can learn from the previous bilateral negotiation sessions it was involved in. In contrast, the research challenge was to make the right decisions to maximize the overall profit in a supply chain environment, such as determining with whom and when to negotiate. This chapter describes the overview of ANL and SCML in ANAC2022, and reports the results of each league, respectively.
Metadata only
A 2020 perspective on “A generalized stereotype learning approach and its instantiation in trust modeling”
(Elsevier, 2020-03) Fang, H.; Zhang, J.; Şensoy, Murat; Computer Science; ŞENSOY, Murat
Owing to the rapid increase of user data and development of machine learning techniques, user modeling has been explored in depth and exploited by both academia and industry. It has prominent impacts in e-commercerelated applications by facilitating users' experience in online platforms and supporting business organizations' decision-making. Among all the techniques and applications, user profiling and recommender systems are two representative and effective ones, which have also obtained growing attention. In view of its wide applications, researchers and practitioners should improve user modeling from two perspectives: (1) more effort should be devoted to obtain more user data via techniques like sensing devices and develop more effective ways to manage complex data; and (2) improving the ability of learning from a limited number of data samples (e.g., few-shot learning) has become an increasingly hot topic for researchers.
Open Access
ACNMP: skill transfer and task extrapolation through learning from demonstration and reinforcement learning via representation sharing
(ML Research Press, 2020) Akbulut, M. T.; Öztop, Erhan; Xue, H.; Tekden, A. E.; Şeker, M. Y.; Uğur, E.; Computer Science; ÖZTOP, Erhan
To equip robots with dexterous skills, an effective approach is to first transfer the desired skill via Learning from Demonstration (LfD), then let the robot improve it by self-exploration via Reinforcement Learning (RL). In this paper, we propose a novel LfD+RL framework, namely Adaptive Conditional Neural Movement Primitives (ACNMP), that allows efficient policy improvement in novel environments and effective skill transfer between different agents. This is achieved through exploiting the latent representation learned by the underlying Conditional Neural Process (CNP) model, and simultaneous training of the model with supervised learning (SL) for acquiring the demonstrated trajectories and via RL for new trajectory discovery. Through simulation experiments, we show that (i) ACNMP enables the system to extrapolate to situations where pure LfD fails; (ii) Simultaneous training of the system through SL and RL preserves the shape of demonstrations while adapting to novel situations due to the shared representations used by both learners; (iii) ACNMP enables order-of-magnitude sample-efficient RL in extrapolation of reaching tasks compared to the existing approaches; (iv) ACNMPs can be used to implement skill transfer between robots having different morphology, with competitive learning speeds and importantly with less number of assumptions compared to the state-of-the-art approaches. Finally, we show the real-world suitability of ACNMPs through real robot experiments that involve obstacle avoidance, pick and place and pouring actions.
Metadata only
Action and language mechanisms in the brain: data, models and neuroinformatics
(Springer Science+Business Media, 2014-01) Arbib, M. A.; Bonaiuto, J. J.; Bornkessel-Schlesewsky, I.; Kemmerer, D.; MacWhinney, B.; Årup Nielsen, F.; Öztop, Erhan; Computer Science; ÖZTOP, Erhan
We assess the challenges of studying action and language mechanisms in the brain, both singly and in relation to each other to provide a novel perspective on neuroinformatics, integrating the development of databases for encoding – separately or together – neurocomputational models and empirical data that serve systems and cognitive neuroscience.
Open Access
Actor-critic reinforcement learning for bidding in bilateral negotiation
(TÜBİTAK, 2022) Arslan, Furkan; Aydoğan, Reyhan; Computer Science; AYDOĞAN, Reyhan; Arslan, Furkan
Designing an effective and intelligent bidding strategy is one of the most compelling research challenges in automated negotiation, where software agents negotiate with each other to find a mutual agreement when there is a conflict of interests. Instead of designing a hand-crafted decision-making module, this work proposes a novel bidding strategy adopting an actor-critic reinforcement learning approach, which learns what to offer in a bilateral negotiation. An entropy reinforcement learning framework called Soft Actor-Critic (SAC) is applied to the bidding problem, and a self-play approach is employed to train the model. Our model learns to produce the target utility of the coming offer based on previous offer exchanges and remaining time. Furthermore, an imitation learning approach called behavior cloning is adopted to speed up the learning process. Also, a novel reward function is introduced that does take not only the agent’s own utility but also the opponent’s utility at the end of the negotiation. The developed agent is empirically evaluated. Thus, a large number of negotiation sessions are run against a variety of opponents selected in different domains varying in size and opposition. The agent’s performance is compared with its opponents and the performance of the baseline agents negotiating with the same opponents. The empirical results show that our agent successfully negotiates against challenging opponents in different negotiation scenarios without requiring any former information about the opponent or domain in advance. Furthermore, it achieves better results than the baseline agents regarding the received utility at the end of the successful negotiations.
Metadata only
Adaptive domain-specific service monitoring
(Springer Science+Business Media, 2014) Ünsal, A. A.; Sazara, G.; Aktemur, Tankut Barış; Sözer, Hasan; Computer Science; AKTEMUR, Tankut Bariş; SÖZER, Hasan
We propose an adaptive and domain-specific service monitoring approach to detect partner service errors in a cost-effective manner. Hereby, we not only consider generic errors such as file not found or connection timed out, but also take domain-specific errors into account. The detection of each type of error entails a different monitoring cost in terms of the consumed resources. To reduce costs, we adapt the monitoring frequency for each service and for each type of error based on the measured error rates and a cost model. We introduce an industrial case study from the broadcasting and content-delivery domain for improving the user-perceived reliability of Smart TV systems. We demonstrate the effectiveness of our approach with real data collected to be relevant for a commercial TV portal application. We present empirical results regarding the trade-off between monitoring overhead and error detection accuracy. Our results show that each service is usually subject to various types of errors with different error rates and exploiting this variation can reduce monitoring costs by up to 30% with negligible compromise on the quality of monitoring.
Metadata only
Adaptive inverse kinematics of a 9-DOF surgical robot for effective manipulation
(IEEE, 2019) Sunal, Begüm; Öztop, Erhan; Bebek, Özkan; Computer Science; Mechanical Engineering; ÖZTOP, Erhan; BEBEK, Özkan; Sunal, Begüm
In a robotic-assisted surgical system, fine and precise movement is essential. However, when the user wishes to cover a wider area during tele-operation, the configuration designed for precise motion may restrict the user and/or slow down the system's operation. This paper proposes a kinematics based method applicable to redundant manipulators to allow the user to make both fast and precise movements and reduce the burden on the user. In the proposed method, different kinematic configurations are selected automatically in real-time to adjust the speed of the robot's end-effector according to the velocity of the haptic device. The proposed method is tested on a 9 Degree-of-Freedom (DOF) system that is realized by attaching a 3-DOF servo driven surgical instrument to a 6-DOF manipulator through a custom interface. The validity of the proposed method is shown with experiments requiring dexterous manipulation using the 9DOF system. The results indicate that adoption of the proposed method in actual operations can facilitate reduction in surgery time and surgeon's effort, thereby help reduce the risk of tissue deformations and other complications in the patient.
Metadata only
Adaptive shared control with human intention estimation for human agent collaboration
(IEEE, 2022) Amirshirzad, Negin; Uğur, E.; Bebek, Özkan; Öztop, Erhan; Computer Science; Mechanical Engineering; BEBEK, Özkan; ÖZTOP, Erhan; Amirshirzad, Negin
In this paper an adaptive shared control frame-work for human agent collaboration is introduced. In this framework the agent predicts the human intention with a confidence factor that also serves as the control blending parameter, that is used to combine the human and agent control commands to drive a robot or a manipulator. While performing a given task, the blending parameter is dynamically updated as the result of the interplay between human and agent control. In a scenario where additional trajectories need to be taught to the agent, either new human demonstrations can be generated and given to the learning system, or alternatively the aforementioned shared control system can be used to generate new demonstrations. The simulation study conducted in this study shows that the latter approach is more beneficial. The latter approach creates improved collaboration between the human and the agent, by decreasing the human effort and increasing the compatibility of the human and agent control commands.
Metadata only
Adaptive streaming of content-aware-encoded videos in dash.js
(IEEE, 2022-05) Beğen, Ali Cengiz; Akçay, Mehmet Necmettin; Bentaleb, A.; Giladi, A.; Computer Science; BEĞEN, Ali Cengiz; Akçay, Mehmet Necmettin
In Hypertext Transfer Protocol (HTTP) adaptive streaming, the client makes rate adaptation decisions based on the measured network bandwidth and buffer fullness. This simplifies the adaptation logic; however, it often produces noticeable quality fluctuations during the streaming session. With content-aware encoding (CAE), one can improve the visual quality without increasing the total number of bits spent by carefully choosing where the bits are spent based on human perception. However, an adaptation logic that is unaware of the resulting variable-size segments may cause more stalls, defeating the purpose of improving viewer experience through CAE. This article explains the design steps of a size-aware rate adaptation (SARA) logic for one of the most popular Dynamic Adaptive Streaming over HTTP (DASH) clients, namely dash.js, and shows the improvements in rebuffering behavior and fetching top-resolution segments as a result of applying this logic.
Metadata only
Adjusting content work flow infrastructures for HDR
(IEEE, 2018-11-28) Syed, Y.; Beğen, Ali Cengiz; Computer Science; BEĞEN, Ali Cengiz
This paper discusses what areas of a content workflow infrastructure are affected by integrating new high dynamic range (HDR) video formats into the system while also still continuing to process the standard dynamic range (SDR) format of the content. First, we present the history of SDR and what benefits HDR can bring. We then look at potential efficiencies in the capture-production-distribution domains where the HDR/SDR paths do not need to be separated and other places where they should be separated to avoid a potential loss in quality. Lastly, we discuss some of the HDR variants and how the infrastructure could handle dynamic metadata as it passes through the relevant workflow domains.
Metadata only
Advancing humanoid robots for social integration: Evaluating trustworthiness through a social cognitive framework
(IEEE, 2023) Taliaronak, V.; Lange, A. L.; Kırtay, M.; Öztop, Erhan; Computer Science; ÖZTOP, Erhan
Trust is an essential concept for human-human and human-robot interactions. Yet only a few studies have addressed this concept from a robot perspective - that is, forming robot trust in interaction partners. Our previous robot trust model relies on assessing the trustworthiness of the interaction partners based on the computational cognitive load incurred during the interactive task [1]. However, this model does not take into account the social markers indicative of trustworthiness, such as the gestures displayed by a human partner. In this study, we make a step toward this point by extending the model by integrating a social cue processing module to achieve social human-robot interaction. This new model serves as a novel social cognitive trust framework to enable the Pepper robot to evaluate the trustworthiness of its interaction partners based on both cognitive load (i.e., the cost of perceptual processing) and social cues (i.e., their gestures). For evaluating the efficacy of the framework, the Pepper robot with the developed model is put to interact with human partners who may take the roles of a reliable, unreliable, deceptive, or random suggestion providing partner. Overall, the results indicate that the proposed framework allows the Pepper robot to differentiate the guiding strategies of the partners by detecting deceptive partners and thus select a trustworthy partner in case of a free choice to perform the next task.
Metadata only
ADVISOR: An adjustable framework for test oracle automation of visual output systems
(IEEE, 2020-09) Genç, A. E.; Sözer, Hasan; Kıraç, Mustafa Furkan; Aktemur, Tankut Barış; Computer Science; SÖZER, Hasan; KIRAÇ, Mustafa Furkan; AKTEMUR, Tankut Bariş
Test oracles differentiate between the correct and incorrect system behavior. Automation of test oracles for visual output systems mainly involves image comparison, where a snapshot of the output is compared with respect to a reference image. Hereby, the captured snapshot can be subject to variations such as scaling and shifting. These variations lead to incorrect evaluations. Existing approaches employ computer vision techniques to address a specific set of variations. In this article, we introduce ADVISOR, an adjustable framework for test oracle automation of visual output systems. It allows the use of a flexible combination and configuration of computer vision techniques. We evaluated a set of valid configurations with a benchmark dataset collected during the tests of commercial digital TV systems. Some of these configurations achieved up to 3% better overall accuracy with respect to state-of-the-art tools. Further, we observed that there is no configuration that reaches the best accuracy for all types of image variations. We also empirically investigated the impact of significant parameters. One of them is a threshold regarding image matching score that determines the final verdict. This parameter is automatically tuned by offline training. We evaluated runtime performance as well. Results showed that differences among the ADVISOR configurations and state-of-the-art tools are in the order of seconds per image comparison.
Open Access
Affordance-based altruistic robotic architecture for human–robot collaboration
(Sage, 2019-08) Imre, M.; Öztop, Erhan; Nagai, Y.; Ugur, E.; Computer Science; ÖZTOP, Erhan
This article proposes a computational model for altruistic behavior, shows its implementation on a physical robot, and presents the results of human-robot interaction experiments conducted with the implemented system. Inspired from the sensorimotor mechanisms of the primate brain, object affordances are utilized for both intention estimation and action execution, in particular, to generate altruistic behavior. At the core of the model is the notion that sensorimotor systems developed for movement generation can be used to process the visual stimuli generated by actions of the others, infer the goals behind, and take the necessary actions to help achieving these goals, potentially leading to the emergence of altruistic behavior. Therefore, we argue that altruistic behavior is not necessarily a consequence of deliberate cognitive processing but may emerge through basic sensorimotor processes such as error minimization, that is, minimizing the difference between the observed and expected outcomes. In the model, affordances also play a key role by constraining the possible set of actions that an observed actor might be engaged in, enabling a fast and accurate intention inference. The model components are implemented on an upper-body humanoid robot. A set of experiments are conducted validating the workings of the components of the model, such as affordance extraction and task execution. Significantly, to assess how human partners interact with our altruistic model deployed robot, extensive experiments with naive subjects are conducted. Our results indicate that the proposed computational model can explain emergent altruistic behavior in reference to its biological counterpart and moreover engage human partners to exploit this behavior when implemented on an anthropomorphic robot.
Metadata only
Agent-based semantic collaborative search
(Executive Committee, Taiwan Academic Network, Ministry of Education, 2013) Şensoy, Murat; Computer Science; ŞENSOY, Murat
Next generation of the Web builds upon technologies such as Semantic Web and Intelligent Software Agents. These technologies aim at knowledge representation that allows both humans and software agents to understand and reason about the content on the Web. In this paper, we propose an agent-based approach for collaborative distributed semantic search of the Web resources. Our approach enables a human user to semantically describe his search interest to an agent. Depending on the interests of their users, the agents evolve their ontologies and create search concepts. Based on these search concepts, the agents coordinate and compose virtual communities. Within these communities, agents with similar interests interact to locate and share URLs relevant to search interests of their users. Through these interactions, shared vocabularies are cooperatively emerged by agents to communicate properly within the communities. Our empirical evaluations and analysis of the proposed approach show that our approach combines Semantic Web technologies and multi-agent systems in a novel way to enable users to find and share the URLs relevant to their search interests.
Metadata only
Agilely assigning sensing assets to mission tasks in a coalition context
(IEEE, 2013) Preece, A.; Norman, T.; de Mel, G.; Pizzocaro, D.; Şensoy, Murat; Pham, T.; Computer Science; ŞENSOY, Murat
When managing intelligence, surveillance, and reconnaissance (ISR) operations in a coalition context, assigning available sensing assets to mission tasks can be challenging. The authors' approach to ISR asset assignment uses ontologies, allocation algorithms, and a service-oriented architecture.
Metadata only
AhBuNe agent: Winner of the eleventh international automated negotiating agent competition (ANAC 2020)
(Springer, 2023) Yıldırım, Ahmet Burak; Sunman, Nezih; Aydoğan, Reyhan; Yıldırım, Ahmet Burak; Sunman, Nezih
The International Automated Negotiating Agent Competition introduces a new challenge each year to facilitate the research on agent-based negotiation and provide a test benchmark. ANAC 2020 addressed the problem of designing effective agents that do not know their users’ complete preferences in addition to their opponent’s negotiation strategy. Accordingly, this paper presents the negotiation strategy of the winner agent called “AhBuNe Agent”. The proposed heuristic-based bidding strategy checks whether it has sufficient orderings to reason about its complete preferences and accordingly decides whether to sacrifice some utility in return for preference elicitation. While making an offer, it uses the most-desired known outcome as a reference and modifies the content of the bid by adopting a concession-based strategy. By analyzing the content of the given ordered bids, the importance ranking of the issues is estimated. As our agent adopts a fixed time-based concession strategy and takes the estimated issue importance ranks into account, it determines to what extent the issues are to be modified. The evaluation results of the ANAC 2020 show that our agent beats the other participating agents in terms of the received individual score.
Metadata only
AIM 2022 challenge on instagram filter removal: Methods and results
(Springer, 2023) Kınlı, Osman Furkan; Menteş, Sami; Özcan, Barış; Kıraç, Mustafa Furkan; Computer Science; KINLI, Osman Furkan; KIRAÇ, Mustafa Furkan; Menteş, Sami; Özcan, Barış
This paper introduces the methods and the results of AIM 2022 challenge on Instagram Filter Removal. Social media filters transform the images by consecutive non-linear operations, and the feature maps of the original content may be interpolated into a different domain. This reduces the overall performance of the recent deep learning strategies. The main goal of this challenge is to produce realistic and visually plausible images where the impact of the filters applied is mitigated while preserving the content. The proposed solutions are ranked in terms of the PSNR value with respect to the original images. There are two prior studies on this task as the baseline, and a total of 9 teams have competed in the final phase of the challenge. The comparison of qualitative results of the proposed solutions and the benchmark for the challenge are presented in this report.
Open Access
Aktör tabanlı sistemler için test kapsama kriterleri
(CEUR-WS, 2018) Sözer, Hasan; Gürler, O.; Yılmaz, O.; Computer Science; Tarhan, A.; Erten, M.; SÖZER, Hasan
Aktör tabanlı sistemler, eşzamanlı çalışan ve birbirleri ile asenkron bir şekilde haberleşen aktör isimli otonom elemanlardan oluşmaktadırlar. Asenkron haberleşme sebebiyle aktörler arasında paylaşılan mesajların sıralaması farklılık gösterebilmektedir. Diğer eşzamanlı çalışan sistemlerde olduğu gibi, determinizm yokluğu, aktör tabanlı sistemlerde test ve hata ayıklama süreçlerini zorlaştırmaktadır. Geleneksel test kapsama kriterleri de bu sistemler için etkin olmamaktadır. Bu bildiride, aktör tabanlı sistemler için kullanılabilecek test kapsama kriterleri irdelenmektedir. Literatürde önerilmiş mevcut kriterlere ek olarak yeni kriterler ile, ayrık matematik alanında son zamanlarda yayınlanmış olan çalışmaların aktör tabanlı sistemler için test kapsama kriterleri belirlemek ve değerlendirmek üzere uygulamaları ilk defa bu bildiride ele alınmaktadır. Önerilen kapsama kriterlerine göre değerlendirme yapabilen ve bu kriterleri sağlayacak şekilde test durumlarını otomatik olarak oluşturan bir test altyapısı şu an geliştirme aşamasındadır.
Open Access
Alarm sequence rule mining extended with a time confidence parameter
(2014) Çelebi, Ö. F.; Zeydan, E.; Arı, İsmail; İleri, Ö.; Ergüt, S.; Computer Science; ARI, Ismail
Most mobile telecommunication operators receive an overwhelming number of alarms in their networks. Network support specialists are faced with the challenge of picking the most important alarms in advance that can cause severe damages to the system or disrupt the service. A system that can discover alarm correlations and alarm rules then notify network administrators can significantly increase the efficiency of Network Operation Centers (NOC) of these mobile operators. This paper provides a new alarm correlation, rule discovery, and significant rule selection technique based on analysis of real data collected from a mobile telecom operator. We present a method based on sequential rule mining algorithm with an additional parameter called time-confidence. The time-confident rules found by this method are processed more efficiently in real-time Complex Event Processing (CEP) systems that require exact time-window values during monitoring. Furthermore, compared to traditional sequential rule mining, our proposed method adds another support dimension to eliminate meaningless rules that appear due to wrong settings of minimum support-confidence thresholds with respect to the nature of data.
Metadata only
Algorithm selection and combining multiple learners for residential energy prediction
(Elsevier, 2019-10) Güngör, Onat; Akşanlı, B.; Aydoğan, Reyhan; Computer Science; AYDOĞAN, Reyhan; Güngör, Onat
Balancing supply and demand management in energy grids requires knowing energy consumption in advance. Therefore, forecasting residential energy consumption accurately plays a key role for future energy systems. For this purpose, in the literature a number of prediction algorithms have been used. This work aims to increase the accuracy of those predictions as much as possible. Accordingly, we first introduce an algorithm selection approach, which identifies the best prediction algorithm for the given residence with respect to its characteristics such as number of people living, appliances and so on. In addition to this, we also study combining multiple learners to increase the accuracy of the predictions. In our experimental setup, we evaluate the aforementioned approaches. Empirical results show that adopting an algorithm selection approach performs better than any single prediction algorithm. Furthermore, combining multiple learners increases the accuracy of the energy consumption prediction significantly.

Browse

Browsing by Title