| 小林 雄太 | 1461003: D, 中間発表 | 松本 裕治，池田 和司，新保 仁，進藤 裕之 |
|
title: Word representation learning with random shuffled texts
abstract: This presentation deals with the learning for distributed vector representations of words. Previous word representation researches consider the bias problem of frequency words with removing them from corpus or adding negative samples generated from unigram distribution to loss function. However, these methods increace the parameters and difficulty of parameter tuning. Starting from an empirical analysis of a word analogy task, we analyze the differences between the properties of word representation which comes from an original texts and a random shuffled texts to extend solutions. language of the presentation: Japanese | ||
| 重藤 優太郎 | 1461004: D, 中間発表 | 松本 裕治，池田 和司，新保 仁，進藤 裕之 |
|
title: Ridge Regression, Hubness, and Zero-Shot Learning
abstract: This presentation discusses the effect of hubness in zero-shot learning, when ridge regression is used to find a mapping between the example space to the label space. Contrary to the existing approach, which attempts to find a mapping from the example space to the label space, we show that mapping labels into the example space is desirable to suppress the emergence of hubs in the subsequent nearest neighbor search step. Assuming a simple data model, we prove that the proposed approach indeed reduces hubness. This was verified empirically on the tasks of bilingual lexicon extraction and image labeling: hubness was reduced with both of these tasks and the accuracy was improved accordingly. language of the presentation: Japanese | ||