Power Loss Function in Neural Networks for Predicting Click-Through Rate

Ergun Biçici. Power Loss Function in Neural Networks for Predicting Click-Through Rate. In Proc. of the 17th ACM Conference on Recommender Systems (RecSys), Singapore, September 2023. URL: https://dl.acm.org/doi/10.1145/3604915.3610658, https://www.growkudos.com/publications/10.1145%252F3604915.3610658/reader

Loss functions guide machine learning models towards concentrating on the error most important to improve upon. We introduce power loss functions for neural networks and apply them on imbalanced click-through rate datasets. Power loss functions decrease the loss for confident predictions and increase the loss for error-prone predictions. They improve both AUC and F1 and produce better calibrated results. We obtain improvements in the results on four different classifiers and on two different datasets. We obtain significant improvements in AUC that reach 0.44% for DeepFM on the Avazu dataset.

Calibrating Neural Networks for CTR Prediction (Reklam Tıklama Oranını Tahmin Etmek için Sinir Ağlarının Kalibrasyonu)

Ergun Biçici and Hasan Saribaş. Calibrating Neural Networks for CTR Prediction (Reklam Tıklama Oranını Tahmin Etmek için Sinir Ağlarının Kalibrasyonu). In Proc. of the 31st Signal Processing and Communications Applications Conference (SIU), Istanbul, Turkey, July 2023. URL: https://ieeexplore.ieee.org/document/10223867

Calibration methods fix the prediction errors of a machine learning model after it is trained and enable more robust and more confident prediction. We implement isotonic regression, Platt's scaling, neural networks, spline regression, and temperature scaling as calibration techniques on the prediction of click-through rate (CTR), which is an unbalanced task. We compare the improvements on using 3 neural network based CTR prediction models, Masknet, DeepFM, and DCNv2, on the publicly available CTR dataset Avazu. Our results demonstrate that isotonic and spline regression methods improve the most and isotonic regression is the fastest method.

Kalibrasyon yöntemleri, bir makine öğrenimi modelinin eğitildikten sonraki tahmin hatalarını düzelterek daha gürbüz ve daha güvenli tahmin yapılmasını sağlar. Bu çalışmada, dengesiz etiketlere sahip bir görev olan reklam tıklama oranının (RTO) tahmininde, kalibrasyon teknikleri olarak izotonik regresyon, Platt'ın ölçeklendirmesi, sinir ağları, eğri regresyonu, ve sıcaklık ölçeklendirmesi uygulanmıştır. Bu kalibrasyon yöntemlerinin karşılaştırılması için halka açık veri seti olan Avazu üzerinde, MaskNet, DeepFM, ve DCNv2 olmak üzere sinir ağı tabanlı üç farklı RTO tahmin modeli kullanılmıştır. Yapılan deneyler, izotonik ve eğri regresyon yöntemlerinin en iyi şekilde performansı arttırdığını ayrıca izotonik regresyonun en hızlı yöntem olduğunu göstermektedir.

Machine Translation Performance Prediction System: Optimal Prediction for Optimal Translation

Ergun Biçici. Machine Translation Performance Prediction System: Optimal Prediction for Optimal Translation. Springer Nature Computer Science, 3, 2022. ISSN: 2661-8907. [doi:10.1007/s42979-022-01183-0]

Machine translation performance prediction (MTPP) system (MTPPS) is an automatic, accurate, language and natural language processing (NLP) output independent prediction model. MTPPS is optimal by the capability to predict translation performance without even using the translation by using only the source, bypassing MT model complexity. MTPPS was casted for tasks involving similarity of text in machine translation (MT), semantic similarity, and parsing of sentences. We present large scale modeling and prediction experiments on MTPP dataset (MTPPDAT) covering $3800$ document- and $380000$ sentence-level prediction in $7$ different domains using $3800$ different MT systems. We provide theoretical and experimental results, empirical lower and upper bounds on the prediction tasks, rank the features used, and present current results. We show that we only need $57$ labeled instances at the document-level and $17$ at the sentence-level to reach current prediction results. MTPPS achieve $4\%$ error rate at the document-level and $45\%$ at the sentence-level relative to the magnitude of the target, $61\%$ and $27\%$ relatively better than a mean predictor correspondingly, and $40\%$ better than the nearest neighbor baseline. Referential translation machines use MTPPS and achieve top results.

RTM Super Learner Results at Quality Estimation Task

 Ergun Biçici. RTM Super Learner Results at Quality Estimation Task. In Proc. of the Sixth Conference on Statistical Machine Translation (WMT21), Online, November 2021.

We obtain new results using referential translation machines (RTMs) with predictions mixed to obtain a better mixture of experts prediction. Our super learner results improve the results and provide a robust combination model.

Parallel Feature Weight Decay Algorithms for Fast Development of Machine Translation Models

Ergun Biçici. Parallel Feature Weight Decay Algorithms for Fast Development of Machine Translation Models. Machine Translation, volume 35, pages 239 - 263, 2021. ISSN: 0922-6567.

Parallel feature weight decay algorithms, parfwd, are engineered for language- and task-adaptive instance selection to build distinct machine translation (MT) models and enable the fast development of accurate MT using fewer data and less computation. parfwd decay the weights of both source and target features to increase their average coverage. In a conference on MT (WMT), parfwd achieved the lowest translation error rate from French to English in 2015, and a rate $11.7\%$ less than the top phrase-based statistical MT (PBSMT) in 2017. parfwd also achieved a rate $5.8\%$ less than the top in TweetMT and the top from Catalan to English. BLEU upper bounds identify the translation directions that offer the largest room for relative improvement and MT models that use additional data. Performance trends angle shows the power of MT models to convert unit data into unit translation results or more BLEU for an increase in coverage. The source coverage angle of parfwd in the 2013--2019 WMT reached +6\textdegree \, better than the top with $35$\textdegree \, for translation into English, and it was +1.4\textdegree \, better than the top with $22$\textdegree \, overall.

RTM Ensemble Learning Results at Quality Estimation Task

Ergun Biçici. RTM Ensemble Learning Results at Quality Estimation Task. In Proc. of the Fifth Conference on Statistical Machine Translation (WMT20), Online, November 2020.

We obtain new results using referential translation machines (RTMs) with predictions mixed and stacked to obtain a better mixture of experts prediction. We are able to achieve better results than the baseline model in Task 1 subtasks. Our stacking results significantly improve the results on the training sets but decrease the test set results. RTMs can achieve to become the 5th among 13 models in ru-en subtask and 5th in the multilingual track of sentence-level Task 1 based on MAE.

8. BÜSİBER Boğaziçi Siber Güvenlik ve KVKK Zirvesi 10.10.2019, 08:29, Albert Long Hall

Boğaziçi Üniversitesi Yönetim Bilişim Sistemleri Siber Güvenlik Merkezi

8. BÜSİBER Boğaziçi Siber Güvenlik ve KVKK Zirvesi

10.10.2019, 08:29, Albert Long Hall
Kişisel Verileri Koruma Kurumu (KVKK, https://www.kvkk.gov.tr/) 24 Mart 2016 6698 sayılı Kişisel Verilerin Korunması Kanunu'na dayanmakta.