0% Complete
فارسی
Home
/
شانزدهمین کنفرانس بین المللی فناوری اطلاعات و دانش
Integrating Wasserstein GANs for High-Speed Transformer-Based Neural Machine Translation
Authors :
Parisa Nekoogol
1
Mostafa Salehi
2
1- دانشگاه تهران
2- دانشگاه تهران
Keywords :
Neural Machine Translation،Generative Adversarial Networks،Reinforcement Learning،Transformer
Abstract :
Neural machine translation (NMT), a key achievement in natural language processing (NLP), continues to face challenges such as producing low-quality output for complex sentences and lacking natural fluency. This study aimed to improve machine translation quality by integrating Generative Adversarial Networks (GANs) with an NMT model. Initially, the baseline NMT model, derived from previous research and based on recurrent neural networks (RNNs), was reconstructed and implemented. Subsequently, this architecture was replaced with the advanced Transformer architecture, and the system was developed using a Wasserstein Generative Adversarial Network (WGAN). To overcome the crucial problem of textual data discontinuity (non-differentiability), the Self-Critical Sequence Training (SCST) method, a reinforcement learning (RL) algorithm, was employed. A core objective was to analyze the performance benefits of adversarial training when applied to a robust Transformer-based generator. The research concluded that while adversarial training enhances the model's performance in generating more fluent translations, this particular improvement is more substantial and notable for models based on recurrent neural networks compared to the Transformer architecture.
Papers List
List of archived papers
طبقهبندی ترافیک رمز مبتنی بر یادگیری ماشین
افسانه معدنی - شقایق نادری - حسین قرایی
HTCAR: Hierarchical Text Classification based on aggregation of Representations
Ali Bavand - Mohammad Mehdi Homayounpour - Ahmad Nickabadi
Open-domain question classification and completion in conversational information search
Omid Mohammadi Kia - Mahmood Neshati - Mahsa Soudi Alamdari
Enhancing Employee Promotion Prediction with a Novel Hybrid Model Integrating Convolutional Neural Networks and Random Forest
Pouya Ardehkhani - Seyyed Reza Moslemi - Hanieh Hooshmand
مدیریت دانش هوشمند مبتنی بر بازیابی-تولید افزوده شده : معماری، ارزیابی و حاکمیت برای دستیار دانش سازمانی
محمدهادی صفری نادری
Detection and Identification of Cyber-Attacks in Cyber-Physical Systems Based on Machine Learning Methods
Zohre Nasiri Zarandi
A method for image steganography based on chaotic maps and advanced compression algorithms
Mohammad Yousefi Sorkhi
تخلیهبار محاسباتی ریزدانه تحرکآگاه در رایانش لبه برای اینترنت اشیاء
شکوفه نوروزی - دکتر زینب موحدی شکوفه نوروزی - زینب موحدی -
Spatial On–Off Keying Modulation with Mirror-Array Optical IRSs for Indoor Machine-to-Machine Visible Light Communication
Babak Sadeghi - Seyed Mohammad Sajad Sadough
ارائه یک مدل تصمیم گیری چند معیاره فازی به منظور بهبود دقت فرایند تصمیم گیری به هنگام اختلال هوانوردی
فاطمه عطا عبدالرزاق - نگار مجمع
more
Samin Hamayesh - Version 42.5.2