Parallelizing Word2Vec in Multi-Core and Many-Core Architectures
This work addresses the problem of computational inefficiency in Word2Vec for researchers and practitioners needing high-speed word embedding extraction, though it is incremental as it builds on existing parallelization methods.
The paper tackled the inefficiency of parallelizing Word2Vec in multi-core architectures by proposing 'HogBatch', which uses minibatching and negative sample sharing to express the problem as matrix multiply operations, achieving the fastest implementation with processing speeds of hundreds of millions of words per second and good scalability up to 32 nodes.
Word2vec is a widely used algorithm for extracting low-dimensional vector representations of words. State-of-the-art algorithms including those by Mikolov et al. have been parallelized for multi-core CPU architectures, but are based on vector-vector operations with "Hogwild" updates that are memory-bandwidth intensive and do not efficiently use computational resources. In this paper, we propose "HogBatch" by improving reuse of various data structures in the algorithm through the use of minibatching and negative sample sharing, hence allowing us to express the problem using matrix multiply operations. We also explore different techniques to distribute word2vec computation across nodes in a compute cluster, and demonstrate good strong scalability up to 32 nodes. The new algorithm is particularly suitable for modern multi-core/many-core architectures, especially Intel's latest Knights Landing processors, and allows us to scale up the computation near linearly across cores and nodes, and process hundreds of millions of words per second, which is the fastest word2vec implementation to the best of our knowledge.