Reinforcement Learning based Curriculum Optimization for Neural Machine Translation

Kumar, Gaurav; Foster, George; Cherry, Colin; Krikun, Maxim

doi:10.18653/v1/n19-1208

Cited by 51 publications

(32 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Platanios et al (2019) propose competence-based curriculum learning that select training samples based on sample difficulty and model competence. Kumar et al (2019) use reinforcement learning to learn the curriculum automatically. propose a norm-based curriculum learning method based on the norm of word embedding to improve the efficiency of training an NMT system.…”

Section: Curriculum Learningmentioning

confidence: 99%

Dynamic Curriculum Learning for Low-Resource Neural Machine Translation

Xu¹,

Hu²,

Jiang³

et al. 2020

Proceedings of the 28th International Conference on Computational Linguistics

View full text Add to dashboard Cite

Large amounts of data has made neural machine translation (NMT) a big success in recent years. But it is still a challenge if we train these models on small-scale corpora. In this case, the way of using data appears to be more important. Here, we investigate the effective use of training data for low-resource NMT. In particular, we propose a dynamic curriculum learning (DCL) method to reorder training samples in training. Unlike previous work, we do not use a static scoring function for reordering. Instead, the order of training samples is dynamically determined in two ways -loss decline and model competence. This eases training by highlighting easy samples that the current model has enough competence to learn. We test our DCL method in a Transformerbased system. Experimental results show that DCL outperforms several strong baselines on three low-resource machine translation benchmarks and different sized data of

show abstract

Section: Curriculum Learningmentioning

confidence: 99%

Dynamic Curriculum Learning for Low-Resource Neural Machine Translation

Xu¹,

Hu²,

Jiang³

et al. 2020

Proceedings of the 28th International Conference on Computational Linguistics

View full text Add to dashboard Cite

show abstract

“…Curriculum Learning In recent years, curriculum learning (Bengio et al, 2009), which enables the models to gradually proceed from easy samples to more complex ones in training (Elman, 1993), has received growing research interests in natural language processing field, e.g., neural machine translation (Platanios et al, 2019;Kumar et al, 2019;Zhao et al, 2020;Liu et al, 2020b;Kocmi and Bojar, 2017;Xu et al, 2020) and computer vision field, e.g., image classification (Weinshall et al, 2018), human attribute analysis and visual question answering (Li et al, 2020). For example, in neural machine translation, Platanios et al (2019) proposed to utilize the training samples in order of easy-to-hard and to describe the "difficulty" of a training sample using the sentence length or the rarity of the words appearing in it (Zhao et al, 2020).…”

Section: Image Captioning and Paragraph Generationmentioning

confidence: 99%

Competence-based Multimodal Curriculum Learning for Medical Report Generation

Liu¹,

Ge²,

Wang³

2021

Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Confer

View full text Add to dashboard Cite

Medical report generation task, which targets to produce long and coherent descriptions of medical images, has attracted growing research interests recently. Different from the general image captioning tasks, medical report generation is more challenging for data-driven neural models. This is mainly due to 1) the serious data bias and 2) the limited medical data. To alleviate the data bias and make best use of available data, we propose a Competencebased Multimodal Curriculum Learning framework (CMCL). Specifically, CMCL simulates the learning process of radiologists and optimizes the model in a step by step manner. Firstly, CMCL estimates the difficulty of each training instance and evaluates the competence of current model; Secondly, CMCL selects the most suitable batch of training instances considering current model competence. By iterating above two steps, CMCL can gradually improve the model's performance. The experiments on the public IU-Xray and MIMIC-CXR datasets show that CMCL can be incorporated into existing models to improve their performance.

show abstract

“…Neural machine translation model training may combine data selection and model training, taking advantage of the increasing quality of the model to better detect noisy data or to increasingly focus on cleaner parts of the data (Wang et al, 2018;Kumar et al, 2019).…”

Section: Impact Of Noise On Neural Machine Translationmentioning

confidence: 99%

Findings of the WMT 2019 Shared Task on Parallel Corpus Filtering for Low-Resource Conditions

Koehn¹,

Guzmán²,

Chaudhary³

et al. 2019

Proceedings of the Fourth Conference on Machine Translation (Volume 3: Shared Task Papers, Day 2)

View full text Add to dashboard Cite

Following the WMT 2018 Shared Task on Parallel Corpus Filtering (Koehn et al., 2018), we posed the challenge of assigning sentencelevel quality scores for very noisy corpora of sentence pairs crawled from the web, with the goal of sub-selecting 2% and 10% of the highest-quality data to be used to train machine translation systems. This year, the task tackled the low resource condition of Nepali-English and Sinhala-English. Eleven participants from companies, national research labs, and universities participated in this task.

show abstract

Reinforcement Learning based Curriculum Optimization for Neural Machine Translation

Cited by 51 publications

References 15 publications

Dynamic Curriculum Learning for Low-Resource Neural Machine Translation

Dynamic Curriculum Learning for Low-Resource Neural Machine Translation

Competence-based Multimodal Curriculum Learning for Medical Report Generation

Findings of the WMT 2019 Shared Task on Parallel Corpus Filtering for Low-Resource Conditions

Contact Info

Product

Resources

About