Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userSelf-Guided Curriculum Learning for Neural Machine Translation
تعلم المناهج ذاتية الذاتية للترجمة الآلية العصبية
405
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
في التعلم الخاضع للإشراف، يجب أن يكون نموذج مدرب جيدا قادرا على استعادة الحقيقة الأرضية بدقة، أي التسميات المتوقعة من المتوقع أن تشبه تسميات الحقيقة الأرضية قدر الإمكان.مستوحاة من ذلك، فإننا صياغة معيارا صعوبة بناء على درجات الاسترداد من أمثلة التدريب.بدافع من الحدس أنه بعد القشط من خلال كوربوس التدريب، يعرف طراز الترجمة الآلية العصبية (NMT) "كيفية جدولة منهج مناسب وفقا لتعلم صعوبة التعلم، نقترح استراتيجية تعلم المناهج الدراسية الموجهة ذاتيا تشجع نموذج NMT للتعلممن سهولة الصعب على أساس درجات الاسترداد.على وجه التحديد، نعتمد درجة بلو على مستوى الجملة باعتبارها وكيل درجة الاسترداد.النتائج التجريبية على معايير الترجمة بما في ذلك WMT14 الإنجليزية والألمانية و WMT17 الصينية - الإنجليزية إظهار أن طريقتنا المقترحة تعمل بشكل كبير على تحسين درجة الاسترداد، وبالتالي تحسين أداء الترجمة باستمرار.
In supervised learning, a well-trained model should be able to recover ground truth accurately, i.e. the predicted labels are expected to resemble the ground truth labels as much as possible. Inspired by this, we formulate a difficulty criterion based on the recovery degrees of training examples. Motivated by the intuition that after skimming through the training corpus, the neural machine translation (NMT) model knows'' how to schedule a suitable curriculum according to learning difficulty, we propose a self-guided curriculum learning strategy that encourages the NMT model to learn from easy to hard on the basis of recovery degrees. Specifically, we adopt sentence-level BLEU score as the proxy of recovery degree. Experimental results on translation benchmarks including WMT14 English-German and WMT17 Chinese-English demonstrate that our proposed method considerably improves the recovery degree, thus consistently improving the translation performance.
References used
https://aclanthology.org/
rate research
Read More
Neural machine translation (NMT) models are data-driven and require large-scale training corpus. In practical applications, NMT models are usually trained on a general domain corpus and then fine-tuned by continuing training on the in-domain corpus.
However, this bears the risk of catastrophic forgetting that the performance on the general domain is decreased drastically. In this work, we propose a new continual learning framework for NMT models. We consider a scenario where the training is comprised of multiple stages and propose a dynamic knowledge distillation technique to alleviate the problem of catastrophic forgetting systematically. We also find that the bias exists in the output linear projection when fine-tuning on the in-domain corpus, and propose a bias-correction module to eliminate the bias. We conduct experiments on three representative settings of NMT application. Experimental results show that the proposed method achieves superior performance compared to baseline models in all settings.
This paper presents Self-correcting Encoding (Secoco), a framework that effectively deals with noisy input for robust neural machine translation by introducing self-correcting predictors. Different from previous robust approaches, Secoco enables NMT
to explicitly correct noisy inputs and delete specific errors simultaneously with the translation decoding process. Secoco is able to achieve significant improvements over strong baselines on two real-world test sets and a benchmark WMT dataset with good interpretability. We will make our code and dataset publicly available soon.
Existing curriculum learning approaches to Neural Machine Translation (NMT) require sampling sufficient amounts of easy'' samples from training data at the early training stage. This is not always achievable for low-resource languages where the amoun
t of training data is limited. To address such a limitation, we propose a novel token-wise curriculum learning approach that creates sufficient amounts of easy samples. Specifically, the model learns to predict a short sub-sequence from the beginning part of each target sentence at the early stage of training. Then the sub-sequence is gradually expanded as the training progresses. Such a new curriculum design is inspired by the cumulative effect of translation errors, which makes the latter tokens more challenging to predict than the beginning ones. Extensive experiments show that our approach can consistently outperform baselines on five language pairs, especially for low-resource languages. Combining our approach with sentence-level methods further improves the performance of high-resource languages.
Neural machine translation (NMT) is sensitive to domain shift. In this paper, we address this problem in an active learning setting where we can spend a given budget on translating in-domain data, and gradually fine-tune a pre-trained out-of-domain N
MT model on the newly translated data. Existing active learning methods for NMT usually select sentences based on uncertainty scores, but these methods require costly translation of full sentences even when only one or two key phrases within the sentence are informative. To address this limitation, we re-examine previous work from the phrase-based machine translation (PBMT) era that selected not full sentences, but rather individual phrases. However, while incorporating these phrases into PBMT systems was relatively simple, it is less trivial for NMT systems, which need to be trained on full sequences to capture larger structural properties of sentences unique to the new domain. To overcome these hurdles, we propose to select both full sentences and individual phrases from unlabelled data in the new domain for routing to human translators. In a German-English translation task, our active learning approach achieves consistent improvements over uncertainty-based sentence selection methods, improving up to 1.2 BLEU score over strong active learning baselines.
Recent research questions the importance of the dot-product self-attention in Transformer models and shows that most attention heads learn simple positional patterns. In this paper, we push further in this research line and propose a novel substitute
mechanism for self-attention: Recurrent AtteNtion (RAN) . RAN directly learns attention weights without any token-to-token interaction and further improves their capacity by layer-to-layer interaction. Across an extensive set of experiments on 10 machine translation tasks, we find that RAN models are competitive and outperform their Transformer counterpart in certain scenarios, with fewer parameters and inference time. Particularly, when apply RAN to the decoder of Transformer, there brings consistent improvements by about +0.5 BLEU on 6 translation tasks and +1.0 BLEU on Turkish-English translation task. In addition, we conduct extensive analysis on the attention weights of RAN to confirm their reasonableness. Our RAN is a promising alternative to build more effective and efficient NMT models.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
