Coping with Zero-Shot Translation and its Explainability

应对零样本翻译及其可解释性

• 批准号: RGPIN-2019-07242
• 负责人: Sadat, Fatiha
• 金额: $ 2.48万
• 依托单位: Université du Québec à Montréal
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=739376
• 关键词: Coping; Zero; Shot; Translation; its

In the last couple of years, the rise of Artificial Intelligence has been remarkable. The data-driven Artificial Intelligence techniques are behind so many of the recent successes and the effects they have on our society. In the field of Natural Language Processing (NLP), a sub-field of Artificial Intelligence, which has close ties with Machine Learning, Neural Machine Translation (NMT) has achieved state-of-the-art translation quality in various research evaluations campaigns when using large amounts of training data. However, such parallel corpora are not widely available for all language pairs and domains. Thus, there is a lot of room for improvement regarding the quality of translation, processing very low-resourced languages and language variants, considering polysynthetic languages, handling the out-of vocabulary, using multiple modalities, translating multiple sentences, , etc. The long-term objective of my research program will deal with the scarcity of algorithm methods for building efficient and transparent zero-shot translation systems and will handle original and challenging problems that give rise to novel algorithms and techniques in the field of machine translation applied to North American indigenous languages.  We intend to contribute to the important challenge of unsupervised machine translation in zero-shot resource settings and an explainable environment, while creating new translation models relying on nothing but monolingual resources and knowledge representations related to this pair of languages. Moreover, we intend to develop new algorithms that explain intermediate outcomes and provide reasoning for the proposed solutions of the automatic translation system, thus unpacking the black box while maintaining a high level of learning performance. The proposed  research Discovery  program will lead to revolutionary advances in the area of machine translation and language resources development and will contribute to the development of innovative technological tools applied to indigenous languages. Moreover, this research program will allow me to consolidate and expand my industrial and international collaborations and will create an excellent research environment for training Highly Qualified Personals in the area of Natural Language Processing and Artificial Intelligence, within three graduate programs in computer science at UQAM.  This research program will consider all forms of diversity to lead to higher-impact research, fosters greater innovation, and results in enhanced performance. As this proposed program will focus on North American indigenous languages as polysynthetic languages, this will help attract talents in indigenous communities and from different ethnic or cultural backgrounds, to strengthen university research, to improve the quality of life for indigenous people and to contribute to stronger economic growth in indigenous communities, and in Canada as a whole.

在过去的几年里，人工智能的崛起引人注目。数据驱动的人工智能技术是最近许多成功的背后，以及它们对我们社会的影响。在与机器学习密切相关的人工智能的子领域自然语言处理（NLP）领域，神经机器翻译（NMT）在使用大量训练数据的各种研究评估活动中取得了最先进的翻译质量。然而，这种平行语料库并不是广泛适用于所有的语言对和语言领域。因此，在翻译质量、处理资源非常少的语言和语言变体、考虑多合成语言、处理词汇不足、使用多种模式、翻译多句等方面还有很大的改进空间。我的研究计划的长期目标将处理构建高效和透明的零射击翻译系统的算法方法的缺乏，并将处理原始和具有挑战性的问题，这些问题将在北美土著语言的机器翻译领域产生新的算法和技术。我们打算在零机会资源设置和可解释环境中为无监督机器翻译的重要挑战做出贡献，同时创建新的翻译模型，仅依赖于单语言资源和与这对语言相关的知识表示。此外，我们打算开发新的算法来解释中间结果，并为自动翻译系统提出的解决方案提供推理，从而在保持高水平学习性能的同时打开黑匣子。拟议的研究发现计划将导致机器翻译和语言资源开发领域的革命性进展，并将有助于开发适用于土著语言的创新技术工具。此外，这个研究项目将使我能够巩固和扩大我的工业和国际合作，并将在UQAM的三个计算机科学研究生课程中为培养自然语言处理和人工智能领域的高素质人才创造一个良好的研究环境。该研究计划将考虑所有形式的多样性，以产生更高影响的研究，促进更大的创新，并提高绩效。由于该计划将重点关注北美土著语言作为多合成语言，这将有助于吸引土著社区和来自不同种族或文化背景的人才，加强大学研究，提高土著人民的生活质量，并为土著社区和整个加拿大的经济增长做出贡献。