Direct Network Transfer: Transfer Learning of Sentence Embeddings for Semantic Similarity
arXiv:1804.07835
Abstract
Sentence encoders, which produce sentence embeddings using neural networks, are typically evaluated by how well they transfer to downstream tasks. This includes semantic similarity, an important task in natural language understanding. Although there has been much work dedicated to building sentence encoders, the accompanying transfer learning techniques have received relatively little attention. In this paper, we propose a transfer learning setting specialized for semantic similarity, which we refer to as direct network transfer. Through experiments on several standard text similarity datasets, we show that applying direct network transfer to existing encoders can lead to state-of-the-art performance. Additionally, we compare several approaches to transfer sentence encoders to semantic similarity tasks, showing that the choice of transfer learning setting greatly affects the performance in many cases, and differs by encoder and dataset.
References in corpus (11)
- Adam: A Method for Stochastic Optimization
- Overcoming catastrophic forgetting in neural networks
- Universal Sentence Encoder
- Unsupervised Learning of Sentence Embeddings using Compositional n-Gram Features
- Understanding Neural Networks through Representation Erasure
- A large annotated corpus for learning natural language inference
- An efficient framework for learning sentence representations
- Advances in Human Action Recognition: A Survey
- Towards Universal Paraphrastic Sentence Embeddings
- How Transferable are Neural Networks in NLP Applications?
- Not All Neural Embeddings are Born Equal