Dani Yogatama

Sparse Overcomplete Word Vector Representations

Current distributed representations of words show little resemblance to theories of lexical semantics. The former are dense and uninterpretable, the latter largely based on familiar, discrete classes (e.g., supersenses) and relations (e.g., synonymy and hypernymy). We propose methods that transform word vectors into sparse (and optionally binary) vectors. The resulting representations are more similar to the interpretable features typically used in NLP, though they are discovered automatically from raw corpora. Because the vectors are highly sparse, they are computationally easy to work with. Most importantly, we find that they outperform the original vectors on benchmark tasks.

Linguistic Structured Sparsity in Text Categorization

We introduce three linguistically motivated structured regularizers based on parse trees, topics, and hierarchical word clusters for text categorization. These regularizers impose linguistic bias in feature weights, enabling us to incorporate prior knowledge into conventional bagof-words models. We show that our structured regularizers consistently improve classification accuracies compared to standard regularizers that penalize features in isolation (such as lasso, ridge, and elastic net regularizers) on a range of datasets for various text prediction problems: topic classification, sentiment analysis, and forecasting.

Embedding Methods for Fine Grained Entity Type Classification

We propose a new approach to the task of fine grained entity type classifications based on label embeddings that allows for information sharing among related labels. Specifically, we learn an embedding for each label and each feature such that labels which frequently co-occur are close in the embedded space. We show that it outperforms state-of-the-art methods on two fine grained entity-classification benchmarks and that the model can exploit the finer-grained labels to improve classification of standard coarse types.

Bayesian Optimization of Text Representations

When applying machine learning to problems in NLP, there are many choices to make about how to represent input texts. These choices can have a big effect on performance, but they are often uninteresting to researchers or practitioners who simply need a module that performs well. We propose an approach to optimizing over this space of choices, formulating the problem as global optimization. We apply a sequential model-based optimization technique and show that our method makes standard linear models competitive with more sophisticated, expensive state-of-the-art methods based on latent variable models or neural networks on various topic classification and sentiment analysis problems. Our approach is a first step towards black-box NLP systems that work with raw text and do not require manual tuning.

Program Induction by Rationale Generation: Learning to Solve and Explain Algebraic Word Problems

Solving algebraic word problems requires executing a series of arithmetic operations---a program---to obtain a final answer. However, since programs can be arbitrarily complicated, inducing them directly from question-answer pairs is a formidable challenge. To make this task more feasible, we solve these problems by generating answer rationales, sequences of natural language and human-readable mathematical expressions that derive the final answer through a series of small steps. Although rationales do not explicitly specify programs, they provide a scaffolding for their structure via intermediate milestones. To evaluate our approach, we have created a new 100,000-sample dataset of questions, answers and rationales. Experimental results show that indirect supervision of program learning via answer rationales is a promising strategy for inducing arithmetic programs.

Extractive Summarization by Maximizing Semantic Volume

The most successful approaches to extractive text summarization seek to maximize bigram coverage subject to a budget constraint. In this work, we propose instead to maximize semantic volume. We embed each sentence in a semantic space and construct a summary by choosing a subset of sentences whose convex hull maximizes volume in that space. We provide a greedy algorithm based on the GramSchmidt process to efficiently perform volume maximization. Our method outperforms the state-of-the-art summarization approaches on benchmark datasets.

Multilingual Spectral Clustering Using Document Similarity Propagation

We present a novel approach for multilingual document clustering using only comparable corpora to achieve cross-lingual semantic interoperability. The method models document collections as weighted graph, and supervisory information is given as sets of must-linked constraints for documents in different languages. Recursive k-nearest neighbor similarity propagation is used to exploit the prior knowledge and merge two language spaces. Spectral method is applied to find the best cuts of the graph. Experimental results show that using limited supervisory information, our method achieves promising clustering results. Furthermore, since the method does not need any language dependent information in the process, our algorithm can be applied to languages in various alphabetical systems.