|Time+Place:||Thursday 16/06/2016 15:00 Room 337-8 Taub Bld.|
|Title:||Understanding Word Embeddings|
|Speaker:|| Omer Levy - CS-Lecture - NOTE UNUSUAL HOUR
|| Affiliation: || Bar-Ilan University
|| Host: || Shaul Markovitch
Neural word embeddings, such as word2vec (Mikolov et al., 2013), have become increasingly popular in both academic and industrial NLP. These methods attempt to capture the semantic meanings of words by processing huge unlabeled corpora with methods inspired by neural networks and the recent onset of Deep Learning. The result is a vectorial representation of every word in a low-dimensional continuous space. These word vectors exhibit interesting arithmetic properties (e.g. king - man + woman = queen) (Mikolov et al., 2013), and seemingly outperform traditional vector-space models of meaning inspired by Harris's Distributional Hypothesis (Baroni et al., 2014). Our work attempts to demystify word embeddings, and understand what makes them so much better than traditional methods at capturing semantic properties. Our main result shows that state-of-the-art word embeddings are actually "more of the same". In particular, we show that skip-grams with negative sampling, the latest algorithm in word2vec, is implicitly factorizing a word-context PMI matrix, which has been thoroughly used and studied in the NLP community for the past 20 years. We also identify that the root of word2vec's perceived superiority can be attributed to a collection of hyperparameter settings. While these hyperparameters were thought to be unique to neural-network-inspired embedding methods, we show that they can, in fact, be ported to traditional distributional methods, significantly improving their performance. Among our qualitative results is a method for interpreting these seemingly-opaque word-vectors, and the answer to why king - man + woman = queen. (Based on joint work with Yoav Goldberg and Ido Dagan.) Short Bio: Omer is a Computer Science PhD student at Bar-Ilan University's Natural Language Processing lab, working with Prof. Ido Dagan and Dr. Yoav Goldberg. He will be starting a post-doc at the University of Washington with Prof. Luke Zettlemoyer this fall. Omer is interested in realizing high-level semantic applications such as question answering and summarization to help people cope with information overload. At the heart of these applications are challenges in textual entailment and semantic similarity, which form the core of Omer's current research. He is also interested in the current advances in representation learning (aka"deep learning"), particularly in the scope of word embeddings, and how they can support semantic applications.