Earth Mover's Distance Minimization for Unsupervised Bilingual
            Lexicon Induction

Zhang, Meng; Liu, Yang; Luan, Huanbo; Sun, Maosong

doi:10.18653/v1/d17-1207

Cited by 109 publications

(107 citation statements)

References 37 publications

Supporting

Mentioning

104

Contrasting

Unclassified

Order By: Relevance

“…There is evidence that different languages represent semantic concepts with similar structure leading to the structural isomorphism across word embedding spaces of different languages [3]. Viewing word embedding spaces as distributions, Zhang et al proposed to build the cross-lingual connection by minimizing their earth movers distance [4]. Differently, they achieve the alignment only by minimizing the distance between distributions while we find the combination of distribution-level and item-level distance minimization can better align the distributions.…”

Section: Introductionmentioning

confidence: 88%

“…where Γ(P 1 , P 2 ) denotes the set of all joint distributions γ(x, y) with marginals P 1 and P 2 . It can be considered as the continuous case of the Earth Mover's Distance, a powerful tool widely used in computer vision and natural language processing [4,8]. Intuitively, if each distribution is viewed as a unit amount of "dirt", earth mover's distance is the minimum "cost" of turning one pile into the other, which is assumed to The framework of SC-GANs.…”

Section: Preliminariesmentioning

confidence: 99%

See 1 more Smart Citation

Semi-Supervised Compatibility Learning Across Categories for Clothing Matching

Zhang

et al. 2019

2019 IEEE International Conference on Multimedia and Expo (ICME)

View full text Add to dashboard Cite

Learning the compatibility between fashion items across categories is a key task in fashion analysis, which can decode the secret of clothing matching. The main idea of this task is to map items into a latent style space where compatible items stay close. Previous works try to build such a transformation by minimizing the distances between annotated compatible items, which require massive item-level supervision. However, these annotated data are expensive to obtain and hard to cover the numerous items with various styles in real applications. In such cases, these supervised methods fail to achieve satisfactory performances. In this work, we propose a semi-supervised method to learn the compatibility across categories. We observe that the distributions of different categories have intrinsic similar structures. Accordingly, the better distributions align, the closer compatible items across these categories become. To achieve the alignment, we minimize the distances between distributions with unsupervised adversarial learning, and also the distances between some annotated compatible items which play the role of anchor points to help align. Experimental results on two real-world datasets demonstrate the effectiveness of our method.

show abstract

Section: Introductionmentioning

confidence: 88%

Section: Preliminariesmentioning

confidence: 99%

Semi-Supervised Compatibility Learning Across Categories for Clothing Matching

Zhang

et al. 2019

2019 IEEE International Conference on Multimedia and Expo (ICME)

View full text Add to dashboard Cite

show abstract

“…For the first application, we have 5 pairs of languages: Chinese-English, Spanish-English, Italian-English, Japanese-Chinese, and Turkish-English. Given by (Zhang et al 2017), each language has a vocabulary list containing 3000 to 13000 words; we also follow their preprocessing idea to represent all the words by vectors in R 50 through the embedding technique (Mikolov, Le, and Sutskever 2013). Actually, each vocabulary list is represented by a distribution in the space where each vector has the weight equal to the corresponding frequency in the language.…”

Section: Methodsmentioning

confidence: 99%

On Geometric Alignment in Low Doubling Dimension

Ding

2019

AAAI

View full text Add to dashboard Cite

In real-world, many problems can be formulated as the alignment between two geometric patterns. Previously, a great amount of research focus on the alignment of 2D or 3D patterns, especially in the field of computer vision. Recently, the alignment of geometric patterns in high dimension finds several novel applications, and has attracted more and more attentions. However, the research is still rather limited in terms of algorithms. To the best of our knowledge, most existing approaches for high dimensional alignment are just simple extensions of their counterparts for 2D and 3D cases, and often suffer from the issues such as high complexities. In this paper, we propose an effective framework to compress the high dimensional geometric patterns and approximately preserve the alignment quality. As a consequence, existing alignment approach can be applied to the compressed geometric patterns and thus the time complexity is significantly reduced. Our idea is inspired by the observation that high dimensional data often has a low intrinsic dimension. We adopt the widely used notion "doubling dimension" to measure the extents of our compression and the resulting approximation. Finally, we test our method on both random and real datasets; the experimental results reveal that running the alignment algorithm on compressed patterns can achieve similar qualities, comparing with the results on the original patterns, but the running times (including the times cost for compression) are substantially lower.

show abstract

“…The amount of required supervision was later reduced through self-learning methods (Artetxe et al, 2017), and then completely eliminated through adversarial training (Zhang et al, 2017a; or more robust iterative approaches combined with initialization heuristics (Artetxe et al, 2018b;Hoshen and Wolf, 2018). At the same time, several recent methods have formulated embedding mappings as an optimal transport problem (Zhang et al, 2017b;.…”

Section: Related Workmentioning

confidence: 99%

Bilingual Lexicon Induction through Unsupervised Machine Translation

Artetxe¹,

Labaka²,

Agirre³

2019

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

View full text Add to dashboard Cite

A recent research line has obtained strong results on bilingual lexicon induction by aligning independently trained word embeddings in two languages and using the resulting crosslingual embeddings to induce word translation pairs through nearest neighbor or related retrieval methods. In this paper, we propose an alternative approach to this problem that builds on the recent work on unsupervised machine translation. This way, instead of directly inducing a bilingual lexicon from cross-lingual embeddings, we use them to build a phrasetable, combine it with a language model, and use the resulting machine translation system to generate a synthetic parallel corpus, from which we extract the bilingual lexicon using statistical word alignment techniques. As such, our method can work with any word embedding and cross-lingual mapping technique, and it does not require any additional resource besides the monolingual corpus used to train the embeddings. When evaluated on the exact same cross-lingual embeddings, our proposed method obtains an average improvement of 6 accuracy points over nearest neighbor and 4 points over CSLS retrieval, establishing a new state-of-the-art in the standard MUSE dataset.

show abstract

Earth Mover's Distance Minimization for Unsupervised Bilingual Lexicon Induction

Cited by 109 publications

References 37 publications

Semi-Supervised Compatibility Learning Across Categories for Clothing Matching

Semi-Supervised Compatibility Learning Across Categories for Clothing Matching

On Geometric Alignment in Low Doubling Dimension

Bilingual Lexicon Induction through Unsupervised Machine Translation

Contact Info

Product

Resources

About