Research Outputs

Permanent URI for this communityhttps://hdl.handle.net/20.500.14288/2

Browse

Filters

Advanced Search

Filter by

Settings

Sponsors: search.filters.sponsors.TÜBİTAKSubject: search.filters.subject.Artificial intelligence

Search Results

Now showing 1 - 10 of 28
  • Placeholder
    PublicationMetadata only
    AffectON: Incorporating affect into dialog generation
    (IEEE-Inst Electrical Electronics Engineers Inc, 2023) Bucinca, Zana; Department of Computer Engineering; Yemez, Yücel; Erzin, Engin; Sezgin, Tevfik Metin; Department of Computer Engineering; Koç Üniversitesi İş Bankası Yapay Zeka Uygulama ve Araştırma Merkezi (KUIS AI)/ Koç University İş Bank Artificial Intelligence Center (KUIS AI); College of Engineering
    Due to its expressivity, natural language is paramount for explicit and implicit affective state communication among humans. The same linguistic inquiry (e.g., How are you?) might induce responses with different affects depending on the affective state of the conversational partner(s) and the context of the conversation. Yet, most dialog systems do not consider affect as constitutive aspect of response generation. In this article, we introduce AffectON, an approach for generating affective responses during inference. For generating language in a targeted affect, our approach leverages a probabilistic language model and an affective space. AffectON is language model agnostic, since it can work with probabilities generated by any language model (e.g., sequence-to-sequence models, neural language models, n-grams). Hence, it can be employed for both affective dialog and affective language generation. We experimented with affective dialog generation and evaluated the generated text objectively and subjectively. For the subjective part of the evaluation, we designed a custom user interface for rating and provided recommendations for the design of such interfaces. The results, both subjective and objective demonstrate that our approach is successful in pulling the generated language toward the targeted affect, with little sacrifice in syntactic coherence.
  • Thumbnail Image
    PublicationOpen Access
    AfriKI: machine-in-the-loop Afrikaans poetry generation
    (Association for Computational Linguistics (ACL), 2021) Baş, Anıl; Department of Comparative Literature; van Heerden, Imke; Other; Department of Comparative Literature; College of Social Sciences and Humanities; 318142
    This paper proposes a generative language model called AfriKI. Our approach is based on an LSTM architecture trained on a small corpus of contemporary fiction. With the aim of promoting human creativity, we use the model as an authoring tool to explore machine-in-the-loop Afrikaans poetry generation. To our knowledge, this is the first study to attempt creative text generation in Afrikaans.
  • Placeholder
    PublicationMetadata only
    Analysis of engagement and user experience with a laughter responsive social robot
    (Isca-int Speech Communication assoc, 2017) N/A; N/A; Department of Computer Engineering; Department of Computer Engineering; Department of Computer Engineering; Türker, Bekir Berker; Buçinca, Zana; Erzin, Engin; Yemez, Yücel; Sezgin, Tevfik Metin; PhD Student; Master Student; Faculty Member; Faculty Member; Faculty Member; Department of Computer Engineering; Graduate School of Sciences and Engineering; Graduate School of Sciences and Engineering; College of Engineering; College of Engineering; College of Engineering; N/A; N/A; 34503; 107907; 18632
    We explore the effect of laughter perception and response in terms of engagement in human-robot interaction. We designed two distinct experiments in which the robot has two modes: laughter responsive and laughter non-responsive. in responsive mode, the robot detects laughter using a multimodal real-time laughter detection module and invokes laughter as a backchannel to users accordingly. in non-responsive mode, robot has no utilization of detection, thus provides no feedback. in the experimental design, we use a straightforward question-answer based interaction scenario using a back-projected robot head. We evaluate the interactions with objective and subjective measurements of engagement and user experience.
  • Placeholder
    PublicationMetadata only
    Audio-visual prediction of head-nod and turn-taking events in dyadic interactions
    (Isca-int Speech Communication assoc, 2018) N/A; Department of Computer Engineering; Department of Computer Engineering; Department of Computer Engineering; Türker, Bekir Berker; Erzin, Engin; Yemez, Yücel; Sezgin, Tevfik Metin; PhD Student; Faculty Member; Faculty Member; Faculty Member; Department of Computer Engineering; Graduate School of Sciences and Engineering; College of Engineering; College of Engineering; College of Engineering; N/A; 34503; 107907; 18632
    Head-nods and turn-taking both significantly contribute conversational dynamics in dyadic interactions. Timely prediction and use of these events is quite valuable for dialog management systems in human-robot interaction. in this study, we present an audio-visual prediction framework for the head-nod and turn taking events that can also be utilized in real-time systems. Prediction systems based on Support vector Machines (SVM) and Long Short-Term Memory Recurrent Neural Networks (LSTM-RNN) are trained on human-human conversational data. Unimodal and multi-modal classification performances of head-nod and turn-taking events are reported over the IEMOCaP dataset.
  • Thumbnail Image
    PublicationOpen Access
    BlockSim-Net: a network-based blockchain simulator
    (TÜBİTAK, 2022) Ramachandran, Prashanthi; Agrawal, Nandini; Department of Computer Engineering; Biçer, Osman; Küpçü, Alptekin; Faculty Member; Department of Computer Engineering; College of Engineering; Graduate School of Sciences and Engineering; N/A; 168060
    Since its proposal by Eyal and Sirer (CACM '13), selfish mining attacks on proof-of-work blockchains have been studied extensively. The main body of this research aims at both studying the extent of its impact and defending against it. Yet, before any practical defense is deployed in a real world blockchain system, it needs to be tested for security and dependability. However, real blockchain systems are too complex to conduct any test on or benchmark the developed protocols. Instead, some simulation environments have been proposed recently, such as BlockSim (Maher et al., SIGMETRICS Perform. Eval. Rev. '19), which is a modular and easy-to-use blockchain simulator. However, BlockSim's structure is insufficient to capture the essence of a real blockchain network, as the simulation of an entire network happens over a single CPU. Such a lack of decentralization can cause network issues such as propagation delays being simulated in an unrealistic manner. In this work, we propose BlockSim-Net, a modular, efficient, high performance, distributed, network-based blockchain simulator that is parallelized to better reflect reality in a blockchain simulation environment.
  • Placeholder
    PublicationMetadata only
    Convolutive bounded component analysis algorithms for independent and dependent source separation
    (IEEE-inst Electrical Electronics Engineers inc, 2015) N/A; N/A; Department of Electrical and Electronics Engineering; İnan, Hüseyin Atahan; Erdoğan, Alper Tunga; Master Student; Faculty Member; Department of Electrical and Electronics Engineering; Graduate School of Sciences and Engineering; College of Engineering; N/A; 41624
    Bounded component analysis (BCa) is a framework that can be considered as a more general framework than independent component analysis (ICa) under the boundedness constraint on sources. Using this framework, it is possible to separate dependent as well as independent components from their mixtures. in this paper, As an extension of a recently introduced instantaneous BCa approach, we introduce a family of convolutive BCa criteria and corresponding algorithms. We prove that the global optima of the proposed criteria, under generic BCa assumptions, Are equivalent to a set of perfect separators. the algorithms introduced in this paper are capable of separating not only the independent sources but also the sources that are dependent/correlated in both component (space) and sample (time) dimensions. therefore, under the condition that the sources are bounded, they can be considered as extended convolutive ICa algorithms with additional dependent/correlated source separation capability. Furthermore, they have potential to provide improvement in separation performance, especially for short data records. This paper offers examples to illustrate the space-time correlated source separation capability through a copula distribution-based example. in addition, A frequency-selective Multiple input Multiple Output equalization example demonstrates the clear performance advantage of the proposed BCa approach over the state-of-the-art ICa-based approaches in setups involving convolutive mixtures of digital communication sources.
  • Placeholder
    PublicationMetadata only
    Diffusion-based isometric depth correspondence
    (Academic Press Inc Elsevier Science, 2019) N/A; N/A; Department of Computer Engineering; Küpçü, Emel; Yemez, Yücel; PhD Student; Faculty Member; Department of Computer Engineering; Graduate School of Sciences and Engineering; College of Engineering; N/A; 107907
    We propose an iterative isometric point correspondence method that relies on diffusion distance to handle challenges posed by commodity depth sensors which usually provide incomplete and noisy surface data exhibiting holes and gaps. We formulate the correspondence problem as finding an optimal partial mapping between two given point sets, that minimizes deviation from isometry. Our algorithm starts with an initial rough correspondence between keypoints, obtained via any point matching technique. This initial correspondence is then pruned and updated by iterating a perfect matching algorithm until convergence in order to find as many reliable correspondences as possible. The resulting set of sparse but reliable correspondences then serves as a base matching from which a dense correspondence set is estimated. We additionally provide a global intrinsic symmetry detection technique which clusters a point cloud into its symmetric sides. We incorporate this technique into our point-based correspondence method so as to address the symmetrical flip problem and to further improve the reliability of our matching results. Our symmetry-aware correspondence method is especially effective on human shapes with global reflectional symmetry. We hence conduct experiments on datasets comprising human shapes and show that our method provides state of the art performance over depth frames exhibiting occlusions, large deformations, and topological noise.
  • Placeholder
    PublicationMetadata only
    Efficient multitask multiple kernel learning with application to cancer research
    (Ieee-Inst Electrical Electronics Engineers Inc, 2022) N/A; N/A; Department of Industrial Engineering; Rahimi, Arezou; Gönen, Mehmet; PhD Student; Faculty Member; Department of Industrial Engineering; Graduate School of Sciences and Engineering; College of Engineering; N/A; 237468
    Multitask multiple kernel learning (MKL) algorithms combine the capabilities of incorporating different data sources into the prediction model and using the data from one task to improve the accuracy on others. However, these methods do not necessarily produce interpretable results. Restricting the solutions to the set of interpretable solutions increases the computational burden of the learning problem significantly, leading to computationally prohibitive run times for some important biomedical applications. That is why we propose a multitask MKL formulation with a clustering of tasks and develop a highly time-efficient solution approach for it. Our solution method is based on the Benders decomposition and treating the clustering problem as finding a given number of tree structures in a graph; hence, it is called the forest formulation. We use our method to discriminate early-stage and late-stage cancers using genomic data and gene sets and compare our algorithm against two other algorithms. The two other algorithms are based on different approaches for linearization of the problem while all algorithms make use of the cutting-plane method. Our results indicate that as the number of tasks and/or the number of desired clusters increase, the forest formulation becomes increasingly favorable in terms of computational performance.
  • Placeholder
    PublicationMetadata only
    Emergency facility location under random network damage: insights from the Istanbul case
    (Pergamon-Elsevier Science Ltd, 2015) Department of Industrial Engineering; N/A; Salman, Fatma Sibel; Yücel, Eda; Faculty Member; PhD Student; Department of Industrial Engineering; College of Engineering; Graduate School of Sciences and Engineering; 178838; 235501
    Damage to infrastructure, especially to highways and roads, adversely affects accessibility to disaster areas. Predicting accessibility to demand points from the supply points by a systematic model would lead to more effective emergency facility location decisions. To this effect, we model the spatial impact of the disaster on network links by random failures with dependency such that failure of a link induces failure of nearby links that are structurally more vulnerable. For each demand point, a set of alternative paths is generated from each potential supply point so that the shortest surviving path will be used for relief transportation after the disaster. The objective is to maximize the expected demand coverage within a specified distance over all possible network realizations. To overcome the computational difficulty caused by extremely large number of possible outcomes, we propose a tabu search heuristic that evaluates candidate solutions over a sample of network scenarios. The scenario generation algorithm that represents the proposed distance and vulnerability based failure model is the main contribution of our study. The tabu search algorithm is applied to Istanbul earthquake preparedness case with a detailed analysis comparing solutions found in no link failure, independent link failure, and dependent link failure cases. The results show that incorporating dependent link failures to the model improves the covered demand percentages significantly.
  • Thumbnail Image
    PublicationOpen Access
    Emotion dependent domain adaptation for speech driven affective facial feature synthesis
    (Institute of Electrical and Electronics Engineers (IEEE), 2022) Department of Electrical and Electronics Engineering; Erzin, Engin; Sadiq, Rizwan; Faculty Member; Department of Electrical and Electronics Engineering; Koç Üniversitesi İş Bankası Yapay Zeka Uygulama ve Araştırma Merkezi (KUIS AI)/ Koç University İş Bank Artificial Intelligence Center (KUIS AI); College of Engineering; 34503; N/A
    Although speech driven facial animation has been studied extensively in the literature, works focusing on the affective content of the speech are limited. This is mostly due to the scarcity of affective audio-visual data. In this article, we improve the affective facial animation using domain adaptation by partially reducing the data scarcity. We first define a domain adaptation to map affective and neutral speech representations to a common latent space in which cross-domain bias is smaller. Then the domain adaptation is used to augment affective representations for each emotion category, including angry, disgust, fear, happy, sad, surprise, and neutral, so that we can better train emotion-dependent deep audio-to-visual (A2V) mapping models. Based on the emotion-dependent deep A2V models, the proposed affective facial synthesis system is realized in two stages: first, speech emotion recognition extracts soft emotion category likelihoods for the utterances; then a soft fusion of the emotion-dependent A2V mapping outputs form the affective facial synthesis. Experimental evaluations are performed on the SAVEE audio-visual dataset. The proposed models are assessed with objective and subjective evaluations. The proposed affective A2V system achieves significant MSE loss improvements in comparison to the recent literature. Furthermore, the resulting facial animations of the proposed system are preferred over the baseline animations in the subjective evaluations.