Recursive deep learning for natural language processing and computer vision

By Richard Socher. 2014

89 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Assessing the Ability of LSTMs to Learn Syntax-Sensitive Dependencies (2016, neutral): “This makes it difficult to distinguish simple but robust sequence models from more expressive architectures (Socher, 2014; Grefenstette et al., 2015; Joulin and Mikolov, 2015).”
A Primer on Neural Network Models for Natural Language Processing (2015, methodology): “The recursive neural network (RecNN) abstraction (Pollack, 1990), popularized in NLP by Richard Socher and colleagues (Socher, Manning, & Ng, 2010; Socher, Lin, Ng, & Manning, 2011; Socher et al., 2013; Socher, 2014) is a generalization of the RNN from sequences to (binary) trees.37 Much like the RN”
Siamese Recurrent Architectures for Learning Sentence Similarity (2016, neutral): “Broader architectural revisions of the sequential RNN have also been introduced for modeling text, including bidirectional, multi-layer (Graves 2012), and recursive tree-structures (Socher 2014).”
Easy-First Dependency Parsing with Hierarchical Tree LSTMs (2016, background): “However, the functions are usually restricted to having a fixed maximum arity (usually two) (Socher et al., 2010; Tai et al., 2015; Socher, 2014).”
Deformation-driven shape correspondence via shape recognition (2017, methodology): “Our model is based on Recursive Autoencoders (RAE) for unlabeled binary trees, developed by Socher et al. (2014). The RAE framework proposed by Socher et al.”
On internal language representations in deep learning: an analysis of machine translation and speech recognition (2018, background): “The composition function can be a single-layer neural network or more sophisticated compositions, and different options have been explored in various tasks [30, 308-312].”