Richard Socher

Glove: Global Vectors for Word Representation

Recent methods for learning vector space representations of words have succeeded in capturing fine-grained semantic and syntactic regularities using vector arithmetic, but the origin of these regularities has remained opaque. We analyze and make explicit the model properties needed for such regularities to emerge in word vectors. The result is a new global logbilinear regression model that combines the advantages of the two major model families in the literature: global matrix factorization and local context window methods. Our model efficiently leverages statistical information by training only on the nonzero elements in a word-word cooccurrence matrix, rather than on the entire sparse matrix or on individual context windows in a large corpus. The model produces a vector space with meaningful substructure, as evidenced by its performance of 75% on a recent word analogy task. It also outperforms related models on similarity tasks and named entity recognition.

Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks

Because of their superior ability to preserve sequence information over time, Long Short-Term Memory (LSTM) networks, a type of recurrent neural network with a more complex computational unit, have obtained strong results on a variety of sequence modeling tasks. The only underlying LSTM structure that has been explored so far is a linear chain. However, natural language exhibits syntactic properties that would naturally combine words to phrases. We introduce the Tree-LSTM, a generalization of LSTMs to tree-structured network topologies. Tree-LSTMs outperform all existing systems and strong LSTM baselines on two tasks: predicting the semantic relatedness of two sentences (SemEval 2014, Task 1) and sentiment classification (Stanford Sentiment Treebank).

A Neural Network for Factoid Question Answering over Paragraphs

Text classification methods for tasks like factoid question answering typically use manually defined string matching rules or bag of words representations. These methods are ineective when question text contains very few individual words (e.g., named entities) that are indicative of the answer. We introduce a recursive neural network (rnn) model that can reason over such input by modeling textual compositionality. We apply our model, qanta, to a dataset of questions from a trivia competition called quiz bowl. Unlike previous rnn models, qanta learns word and phrase-level representations that combine across sentences to reason about entities. The model outperforms multiple baselines and, when combined with information retrieval methods, rivals the best human players.

Connecting modalities: Semi-supervised segmentation and annotation of images using unaligned text corpora

We propose a semi-supervised model which segments and annotates images using very few labeled images and a large unaligned text corpus to relate image regions to text labels. Given photos of a sports event, all that is necessary to provide a pixel-level labeling of objects and background is a set of newspaper articles about this sport and one to five labeled images. Our model is motivated by the observation that words in text corpora share certain context and feature similarities with visual objects. We describe images using visual words, a new region-based representation. The proposed model is based on kernelized canonical correlation analysis which finds a mapping between visual and textual words by projecting them into a latent meaning space. Kernels are derived from context and adjective features inside the respective visual and textual domains. We apply our method to a challenging dataset and rely on articles of the New York Times for textual features. Our model outperforms the state-of-the-art in annotation. In segmentation it compares favorably with other methods that use significantly more labeled training data.

Knowing When to Look: Adaptive Attention via a Visual Sentinel for Image Captioning

Attention-based neural encoder-decoder frameworks have been widely adopted for image captioning. Most methods force visual attention to be active for every generated word. However, the decoder likely requires little to no visual information from the image to predict non-visual words such as the and of. Other words that may seem visual can often be predicted reliably just from the language model e.g., sign after behind a red stop or phone following talking on a cell. In this paper, we propose a novel adaptive attention model with a visual sentinel. At each time step, our model decides whether to attend to the image (and if so, to which regions) or to the visual sentinel. The model decides whether to attend to the image and where, in order to extract meaningful information for sequential word generation. We test our method on the COCO image captioning 2015 challenge dataset and Flickr30K. Our approach sets the new state-of-the-art by a significant margin.

A Joint Many-Task Model: Growing a Neural Network for Multiple NLP Tasks

Transfer and multi-task learning have traditionally focused on either a single source-target pair or very few, similar tasks. Ideally, the linguistic levels of morphology, syntax and semantics would benefit each other by being trained in a single model. We introduce a joint many-task model together with a strategy for successively growing its depth to solve increasingly complex tasks. Higher layers include shortcut connections to lower-level task predictions to reflect linguistic hierarchies. We use a simple regularization term to allow for optimizing all model weights to improve one task’s loss without exhibiting catastrophic interference of the other tasks. Our single end-to-end model obtains state-of-the-art or competitive results on five different tasks from tagging, parsing, relatedness, and entailment tasks.

Scaling short-answer grading by combining peer assessment with algorithmic scoring

Peer assessment helps students reflect and exposes them to different ideas. It scales assessment and allows large online classes to use open-ended assignments. However, it requires students to spend significant time grading. How can we lower this grading burden while maintaining quality? This paper integrates peer and machine grading to preserve the robustness of peer assessment and lower grading burden. In the identify-verify pattern, a grading algorithm first predicts a student grade and estimates confidence, which is used to estimate the number of peer raters required. Peers then identify key features of the answer using a rubric. Finally, other peers verify whether these feature labels were accurately applied. This pattern adjusts the number of peers that evaluate an answer based on algorithmic confidence and peer agreement. We evaluated this pattern with 1370 students in a large, online design class. With only 54% of the student grading time, the identify-verify pattern yields 80-90% of the accuracy obtained by taking the median of three peer scores, and provides more detailed feedback. A second experiment found that verification dramatically improves accuracy with more raters, with a 20% gain over the peer-median with four raters. However, verification also leads to lower initial trust in the grading system. The identify-verify pattern provides an example of how peer work and machine learning can combine to improve the learning experience.

Deep Learning for Sentiment Analysis - Invited Talk.

Richard Socher is the CEO and founder of MetaMind, a startup that seeks to improve artificial intelligence and make it widely accessible. He obtained his PhD from Stanford working on deep learning with Chris Manning and Andrew Ng and won the best Stanford CS PhD thesis award. He is interested in developing new AI models that perform well across multiple different tasks in natural language processing and computer vision. He was awarded the Distinguished Application Paper Award at the International Conference on Machine Learning (ICML) 2011, the 2011 Yahoo! Key Scientific Challenges Award, a Microsoft Research PhD Fellowship in 2012 and a 2013 ”Magic Grant” from the Brown Institute for Media Innovation and the 2014 GigaOM Structure Award.

MetaMind Neural Machine Translation System for WMT 2016.

Neural Machine Translation (NMT) systems, introduced only in 2013, have achieved state of the art results in many MT tasks. MetaMind’s submissions to WMT ’16 seek to push the state of the art in one such task, English→German newsdomain translation. We integrate promising recent developments in NMT, including subword splitting and back-translation for monolingual data augmentation, and introduce the Y-LSTM, a novel neural translation architecture.

Learning when to skim and when to read.

Many recent advances in deep learning for natural language processing have come at increasing computational cost, but the power of these state-of-the-art models is not needed for every example in a dataset. We demonstrate two approaches to reducing unnecessary computation in cases where a fast but weak baseline classier and a stronger, slower model are both available. Applying an AUC-based metric to the task of sentiment classification, we find significant efficiency gains with both a probability-threshold method for reducing computational cost and one that uses a secondary decision network.