Fine-grained Analysis of Sentence Embeddings Using Auxiliary Prediction Tasks

Abstract

There is a lot of research interest in encoding variable length sentences into fixed length vectors, in a way that preserves the sentence meanings. Two common methods include representations based on averaging word vectors, and representations based on the hidden states of recurrent neural networks such as LSTMs. The sentence vectors are used as features for subsequent machine learning tasks or for pre-training in the context of deep learning. However, not much is known about the properties that are encoded in these sentence representations and about the language information they capture. We propose a framework that facilitates better understanding of the encoded representations. We define prediction tasks around isolated aspects of sentence structure (namely sentence length, word content, and word order), and score representations by the ability to train a classifier to solve each prediction task when using the representation as input. We demonstrate the potential contribution of the approach by analyzing different sentence representation mechanisms. The analysis sheds light on the relative strengths of different sentence embedding methods with respect to these low level prediction tasks, and on the effect of the encoded vector's dimensionality on the resulting representations.

References in corpus (4)

• Improving neural networks by preventing co-adaptation of feature detectors
• ADADELTA: An Adaptive Learning Rate Method
• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• A Hierarchical Neural Autoencoder for Paragraphs and Documents

Cited by in corpus (17)

• Pointer Sentinel Mixture Models
• Syntactic Structure from Deep Learning
• Evaluating Layers of Representation in Neural Machine Translation on Part-of-Speech and Semantic Tagging Tasks
• Interpreting Deep Learning Models in Natural Language Processing: A Review
• Pre-Trained Models: Past, Present and Future
• Probing artificial neural networks: insights from neuroscience
• A Further Study of Unsupervised Pre-training for Transformer Based Speech Recognition
• Probing Neural Dialog Models for Conversational Understanding
• Discovery of Natural Language Concepts in Individual Units of CNNs
• Trick or TReAT: Thematic Reinforcement for Artistic Typography
• On the Interplay Between Fine-tuning and Composition in Transformers
• Interpreting the Syntactic and Social Elements of the Tweet Representations via Elementary Property Prediction Tasks
• Probing Task-Oriented Dialogue Representation from Language Models
• Vec2Sent: Probing Sentence Embeddings with Natural Language Generation
• Evaluating the Representational Hub of Language and Vision Models
• Bag-of-Vector Embeddings of Dependency Graphs for Semantic Induction
• RuSentEval: Linguistic Source, Encoder Force!