7 sonuçlar
Arama Sonuçları
Listeleniyor 1 - 7 / 7
Yayın Soft decision trees(IEEE, 2012) İrsoy, Ozan; Yıldız, Olcay Taner; Alpaydın, Ahmet İbrahim EthemWe discuss a novel decision tree architecture with soft decisions at the internal nodes where we choose both children with probabilities given by a sigmoid gating function. Our algorithm is incremental where new nodes are added when needed and parameters are learned using gradient-descent. We visualize the soft tree fit on a toy data set and then compare it with the canonical, hard decision tree over ten regression and classification data sets. Our proposed model has significantly higher accuracy using fewer nodes.Yayın Decoder-side super-resolution and frame interpolation for improved H.264 video coding(IEEE, 2013) Ateş, Hasan FehmiIn literature decoder-side motion estimation is shown to improve video coding efficiency of both H.264 and HEVC standards. In this paper we introduce enhanced skip and direct modes for H.264 coding using decoder-side super-resolution (SR) and frame interpolation. P-and B-frames are downsampled and H.264 encoded at lower resolution (LR). Then reconstructed LR frames are super-resolved using decoder-side motion estimation. Alternatively for B-frames, bidirectional true motion estimation is performed to synthesize a B-frame from its reference frames. For P-frames, bicubic interpolation of the LR frame is used as an alternative to SR reconstruction. A rate-distortion optimal mode selection algorithm determines for each MB which of the two reconstructions to use as skip/direct mode prediction. Simulations indicate an average of 1.04 dB PSNR improvement or 23.0% bitrate reduction at low bitrates when compared to H.264 standard. Average PSNR gains reach as high as 3.95 dB depending on the video content and frame rate.Yayın Çok-hipotezli süperpikseller ile sahne bölütleme ve etiketleme(IEEE, 2015-06-19) Ak, Kenan Emir; Ateş, Hasan FehmiSüperpikseller son zamanlarda imge bölütleme ve sınıflandırma problemlerinde giderek önem kazanmaktadır. Sahne etiketlemede imge öncelikle bir süperpiksel algoritması ile görsel olarak tutarlı küçük parçalara bölütlenmekte; daha sonra süperpikseller farklı sınıflara ayrı¸stırılmaktadır. Sınıflandırma performansı kullanılan süperpiksel algoritmasının özellikleri ve parametre ayarlarından önemli ölçüde etkilenmektedir. Bu bildiride sahne etiketleme doğruluğunu iyileştirmek için birden fazla süperpiksel bölütleme sonucunu sınıflandırıcı seviyesinde kaynaştıran bir yöntem önerilmiştir. Öncelikle basit, parametrik olmayan ve eğitim gerektirmeyen SuperParsing algoritması kullanılarak süperpiksel etiketleri için olabilirlik oranları tespit edilir. Daha sonra alternatif süperpiksel bölütleme senaryoları için hesaplanan olabilirlik oranları piksel seviyesinde kaynaştırılarak, ilgili sahnenin bölütlenmesi ve etiketlenmesi tamamlanır. Önerilen yöntem 2,688 imge ve 33 etiket içeren SIFT Flow veri kümesi üzerinde test edilmiş ve SuperParsing’den daha yüksek sınıflandırma doğruluğu elde edilmiştir.Yayın Budding trees(IEEE Computer Soc, 2014-08-24) İrsoy, Ozan; Yıldız, Olcay Taner; Alpaydın, Ahmet İbrahim EthemWe propose a new decision tree model, named the budding tree, where a node can be both a leaf and an internal decision node. Each bud node starts as a leaf node, can then grow children, but then later on, if necessary, its children can be pruned. This contrasts with traditional tree construction algorithms that only grows the tree during the training phase, and prunes it in a separate pruning phase. We use a soft tree architecture and show that the tree and its parameters can be trained using gradient-descent. Our experimental results on regression, binary classification, and multi-class classification data sets indicate that our newly proposed model has better performance than traditional trees in terms of accuracy while inducing trees of comparable size.Yayın A hybrid approach to private record matching(IEEE Computer Soc, 2012-10) İnan, Ali; Kantarcıoğlu, Murat; Ghinita, Gabriel; Bertino, ElisaReal-world entities are not always represented by the same set of features in different data sets. Therefore, matching records of the same real-world entity distributed across these data sets is a challenging task. If the data sets contain private information, the problem becomes even more difficult. Existing solutions to this problem generally follow two approaches: sanitization techniques and cryptographic techniques. We propose a hybrid technique that combines these two approaches and enables users to trade off between privacy, accuracy, and cost. Our main contribution is the use of a blocking phase that operates over sanitized data to filter out in a privacy-preserving manner pairs of records that do not satisfy the matching condition. We also provide a formal definition of privacy and prove that the participants of our protocols learn nothing other than their share of the result and what can be inferred from their share of the result, their input and sanitized views of the input data sets (which are considered public information). Our method incurs considerably lower costs than cryptographic techniques and yields significantly more accurate matching results compared to sanitization techniques, even when privacy requirements are high.Yayın ANN activation function estimators for homomorphic encrypted inference(Institute of Electrical and Electronics Engineers Inc., 2025-06-13) Harb, Mhd Raja Abou; Çeliktaş, BarışHomomorphic Encryption (HE) enables secure computations on encrypted data, facilitating machine learning inference in sensitive environments such as healthcare and finance. However, efficiently handling non-linear activation functions, specifically Sigmoid and Tanh, remains a significant computational challenge for encrypted inference using Artificial Neural Networks (ANNs). This study introduces a lightweight, ANN-based estimator designed to accurately approximate activation functions under homomorphic encryption. Unlike traditional polynomial and piecewise linear approximations, the proposed ANN estimators achieve superior accuracy with lower computational overhead associated with bootstrapping or high-degree polynomial techniques. These estimators are trained on plaintext data and seamlessly integrated into encrypted inference pipelines, significantly outperforming conventional methods. Experimental evaluations demonstrate notable improvements, with ANN estimators enhancing accuracy by approximately 2% for Sigmoid and up to 73% for Tanh functions, improving F1-scores by approximately 2% for Sigmoid and up to 88% for Tanh, and markedly reducing Mean Square Error (MSE) by up to 96% compared to polynomial approximations. The ANN estimator achieves an accuracy of 97.70% and an AUC of 0.9997 when integrated into a CNN architecture on the MNIST dataset, and an accuracy of 85.25% with an AUC of 0.9459 on the UCI Heart Disease dataset during ciphertext inference. These results underscore the estimator’s practical effectiveness and computational feasibility, making it suitable for secure and efficient ANN inference in encrypted environments.Yayın Grammar or crammer? the role of morphology in distinguishing orthographically similar but semantically unrelated words(Institute of Electrical and Electronics Engineers Inc., 2025) Ercan, Gökhan; Yıldız, Olcay TanerWe show that n-gram-based distributional models fail to distinguish unrelated words due to the noise in semantic spaces. This issue remains hidden in conventional benchmarks but becomes more pronounced when orthographic similarity is high. To highlight this problem, we introduce OSimUnr, a dataset of nearly one million English and Turkish word-pairs that are orthographically similar but semantically unrelated (e.g., grammar - crammer). These pairs are generated through a graph-based WordNet approach and morphological resources. We define two evaluation tasks - unrelatedness identification and relatedness classification - to test semantic models. Our experiments reveal that FastText, with default n-gram segmentation, performs poorly (below 5% accuracy) in identifying unrelated words. However, morphological segmentation overcomes this issue, boosting accuracy to 68% (English) and 71% (Turkish) without compromising performance on standard benchmarks (RareWords, MTurk771, MEN, AnlamVer). Furthermore, our results suggest that even state-of-the-art LLMs, including Llama 3.3 and GPT-4o-mini, may exhibit noise in their semantic spaces, particularly in highly synthetic languages such as Turkish. To ensure dataset quality, we leverage WordNet, MorphoLex, and NLTK, covering fully derivational morphology supporting atomic roots (e.g., '-co_here+ance+y' for 'coherency'), with 405 affixes in Turkish and 467 in English.
