arXiv subjects

Percy Liang

Publications and source records attributed to Percy Liang.

At least 271 records · Page 15Linked to original sources

arXiv2016

Unsupervised Risk Estimation Using Only Conditional Independence Structure

We show how to estimate a model's test error from unlabeled data, on distributions very different from the training distribution, while assuming only that certain conditional independencies are preserved between train and test. We do not need to assume that the optimal predictor is the same between train and test, or that the true distribution lies in any parametric family. We can also efficiently differentiate the error estimate to perform unsupervised discriminative learning. Our technical tool is the method of moments, which allows us to exploit conditional independencies in the absence of a fully-specified model. Our framework encompasses a large family of losses including the log and exponential loss, and extends to structured output settings such as hidden Markov models.

cs.LG↗

arXiv2016

Simpler Context-Dependent Logical Forms via Model Projections

We consider the task of learning a context-dependent mapping from utterances to denotations. With only denotations at training time, we must search over a combinatorially large space of logical forms, which is even larger with context-dependent utterances. To cope with this challenge, we perform successive projections of the full model onto simpler models that operate over equivalence classes of logical forms. Though less expressive, we find that these simpler models are much faster and can be surprisingly effective. Moreover, they can be used to bootstrap the full model. Finally, we collected three new context-dependent semantic parsing datasets, and develop a new left-to-right parser.

cs.CL↗

arXiv2016

Data Recombination for Neural Semantic Parsing

Modeling crisp logical regularities is crucial in semantic parsing, making it difficult for neural models with no task-specific prior knowledge to achieve good results. In this paper, we introduce data recombination, a novel framework for injecting such prior knowledge into a model. From the training data, we induce a high-precision synchronous context-free grammar, which captures important conditional independence properties commonly found in semantic parsing. We then train a sequence-to-sequence recurrent network (RNN) model with a novel attention-based copying mechanism on datapoints sampled from this grammar, thereby teaching the model about these structural properties. Data recombination improves the accuracy of our RNN model on three semantic parsing datasets, leading to new state-of-the-art performance on the standard GeoQuery dataset for models with comparable supervision.

cs.CL↗

arXiv2016

Learning Language Games through Interaction

We introduce a new language learning setting relevant to building adaptive natural language interfaces. It is inspired by Wittgenstein's language games: a human wishes to accomplish some task (e.g., achieving a certain configuration of blocks), but can only communicate with a computer, who performs the actual actions (e.g., removing all red blocks). The computer initially knows nothing about language and therefore must learn it from scratch through interaction, while the human adapts to the computer's capabilities. We created a game in a blocks world and collected interactions from 100 people playing it. First, we analyze the humans' strategies, showing that using compositionality and avoiding synonyms correlates positively with task performance. Second, we compare computer strategies, showing how to quickly learn a semantic parsing model from scratch, and that modeling pragmatics further accelerates learning for successful players.

cs.CL↗

arXiv2016

Estimating Mixture Models via Mixtures of Polynomials

Mixture modeling is a general technique for making any simple model more expressive through weighted combination. This generality and simplicity in part explains the success of the Expectation Maximization (EM) algorithm, in which updates are easy to derive for a wide class of mixture models. However, the likelihood of a mixture model is non-convex, so EM has no known global convergence guarantees. Recently, method of moments approaches offer global guarantees for some mixture models, but they do not extend easily to the range of mixture models that exist. In this work, we present Polymom, an unifying framework based on method of moments in which estimation procedures are easily derivable, just as in EM. Polymom is applicable when the moments of a single mixture component are polynomials of the parameters. Our key observation is that the moments of the mixture model are a mixture of these polynomials, which allows us to cast estimation as a Generalized Moment Problem. We solve its relaxations using semidefinite optimization, and then extract parameters using ideas from computer algebra. This framework allows us to draw insights and apply tools from convex optimization, computer algebra and the theory of moments to study problems in statistical estimation.

stat.ML↗

arXiv2016

Learning Executable Semantic Parsers for Natural Language Understanding

For building question answering systems and natural language interfaces, semantic parsing has emerged as an important and powerful paradigm. Semantic parsers map natural language into logical forms, the classic representation for many important linguistic phenomena. The modern twist is that we are interested in learning semantic parsers from data, which introduces a new layer of statistical and computational issues. This article lays out the components of a statistical semantic parser, highlighting the key challenges. We will see that semantic parsing is a rich fusion of the logical and the statistical world, and that this fusion will play an integral role in the future of natural language understanding systems.

cs.CL↗

arXiv2016

Data Augmentation via Levy Processes

If a document is about travel, we may expect that short snippets of the document should also be about travel. We introduce a general framework for incorporating these types of invariances into a discriminative classifier. The framework imagines data as being drawn from a slice of a Levy process. If we slice the Levy process at an earlier point in time, we obtain additional pseudo-examples, which can be used to train the classifier. We show that this scheme has two desirable properties: it preserves the Bayes decision boundary, and it is equivalent to fitting a generative model in the limit where we rewind time back to 0. Our construction captures popular schemes such as Gaussian feature noising and dropout training, as well as admitting new generalizations.

stat.ML↗

arXiv2015

On-the-Job Learning with Bayesian Decision Theory

Our goal is to deploy a high-accuracy system starting with zero training examples. We consider an "on-the-job" setting, where as inputs arrive, we use real-time crowdsourcing to resolve uncertainty where needed and output our prediction when confident. As the model improves over time, the reliance on crowdsourcing queries decreases. We cast our setting as a stochastic game based on Bayesian decision theory, which allows us to balance latency, cost, and accuracy objectives in a principled way. Computing the optimal policy is intractable, so we develop an approximation based on Monte Carlo Tree Search. We tested our approach on three datasets---named-entity recognition, sentiment classification, and image classification. On the NER task we obtained more than an order of magnitude reduction in cost compared to full human annotation, while boosting performance relative to the expert provided labels. We also achieve a 8% F1 improvement over having a single human label the whole set, and a 28% F1 improvement over online learning.

cs.AI↗

arXiv2015

Traversing Knowledge Graphs in Vector Space

Path queries on a knowledge graph can be used to answer compositional questions such as "What languages are spoken by people living in Lisbon?". However, knowledge graphs often have missing facts (edges) which disrupts path queries. Recent models for knowledge base completion impute missing facts by embedding knowledge graphs in vector spaces. We show that these models can be recursively applied to answer path queries, but that they suffer from cascading errors. This motivates a new "compositional" training objective, which dramatically improves all models' ability to answer path queries, in some cases more than doubling accuracy. On a standard knowledge base completion task, we also demonstrate that compositional training acts as a novel form of structural regularization, reliably improving performance across all base models (reducing errors by up to 43%) and achieving new state-of-the-art results.

cs.CL↗

arXiv2015

Compositional Semantic Parsing on Semi-Structured Tables

Two important aspects of semantic parsing for question answering are the breadth of the knowledge source and the depth of logical compositionality. While existing work trades off one aspect for another, this paper simultaneously makes progress on both fronts through a new task: answering complex questions on semi-structured tables using question-answer pairs as supervision. The central challenge arises from two compounding factors: the broader domain results in an open-ended set of relations, and the deeper compositionality results in a combinatorial explosion in the space of logical forms. We propose a logical-form driven parsing algorithm guided by strong typing constraints and show that it obtains significant improvements over natural baselines. For evaluation, we created a new dataset of 22,033 complex questions on Wikipedia tables, which is made publicly available.

cs.CL↗

arXiv2015

Tensor Factorization via Matrix Factorization

Tensor factorization arises in many machine learning applications, such knowledge base modeling and parameter estimation in latent variable models. However, numerical methods for tensor factorization have not reached the level of maturity of matrix factorization methods. In this paper, we propose a new method for CP tensor factorization that uses random projections to reduce the problem to simultaneous matrix diagonalization. Our method is conceptually simple and also applies to non-orthogonal and asymmetric tensors of arbitrary order. We prove that a small number random projections essentially preserves the spectral information in the tensor, allowing us to remove the dependence on the eigengap that plagued earlier tensor-to-matrix reductions. Experimentally, our method outperforms existing tensor factorization methods on both simulated data and two real datasets.

cs.LG↗

arXiv2015

Simultaneous diagonalization: the asymmetric, low-rank, and noisy settings

Simultaneous matrix diagonalization is used as a subroutine in many machine learning problems, including blind source separation and paramater estimation in latent variable models. Here, we extend algorithms for performing joint diagonalization to low-rank and asymmetric matrices, and we also provide extensions to the perturbation analysis of these methods. Our results allow joint diagonalization to be applied in several new settings.

math.NA↗

arXiv2015

Reified Context Models

A classic tension exists between exact inference in a simple model and approximate inference in a complex model. The latter offers expressivity and thus accuracy, but the former provides coverage of the space, an important property for confidence estimation and learning with indirect supervision. In this work, we introduce a new approach, reified context models, to reconcile this tension. Specifically, we let the amount of context (the arity of the factors in a graphical model) be chosen "at run-time" by reifying it---that is, letting this choice itself be a random variable inside the model. Empirically, we show that our approach obtains expressivity and coverage on three natural language tasks.

cs.LG↗

arXiv2015

Learning Fast-Mixing Models for Structured Prediction

Markov Chain Monte Carlo (MCMC) algorithms are often used for approximate inference inside learning, but their slow mixing can be difficult to diagnose and the approximations can seriously degrade learning. To alleviate these issues, we define a new model family using strong Doeblin Markov chains, whose mixing times can be precisely controlled by a parameter. We also develop an algorithm to learn such models, which involves maximizing the data likelihood under the induced stationary distribution of these chains. We show empirical improvements on two challenging inference tasks.

cs.LG↗

arXiv2014

The Statistics of Streaming Sparse Regression

We present a sparse analogue to stochastic gradient descent that is guaranteed to perform well under similar conditions to the lasso. In the linear regression setup with irrepresentable noise features, our algorithm recovers the support set of the optimal parameter vector with high probability, and achieves a statistically quasi-optimal rate of convergence of Op(k log(d)/T), where k is the sparsity of the solution, d is the number of features, and T is the number of training examples. Meanwhile, our algorithm does not require any more computational resources than stochastic gradient descent. In our experiments, we find that our method substantially out-performs existing streaming algorithms on both real and simulated data.

math.ST↗

arXiv2014

Altitude Training: Strong Bounds for Single-Layer Dropout

Dropout training, originally designed for deep neural networks, has been successful on high-dimensional single-layer natural language tasks. This paper proposes a theoretical explanation for this phenomenon: we show that, under a generative Poisson topic model with long documents, dropout training improves the exponent in the generalization bound for empirical risk minimization. Dropout achieves this gain much like a marathon runner who practices at altitude: once a classifier learns to perform reasonably well on training examples that have been artificially corrupted by dropout, it will do very well on the uncorrupted test set. We also show that, under similar conditions, dropout preserves the Bayes decision boundary and should therefore induce minimal bias in high dimensions.

stat.ML↗

arXiv2014

Relaxations for inference in restricted Boltzmann machines

We propose a relaxation-based approximate inference algorithm that samples near-MAP configurations of a binary pairwise Markov random field. We experiment on MAP inference tasks in several restricted Boltzmann machines. We also use our underlying sampler to estimate the log-partition function of restricted Boltzmann machines and compare against other sampling-based methods.

stat.ML↗

arXiv2013

Dropout Training as Adaptive Regularization

Dropout and other feature noising schemes control overfitting by artificially corrupting the training data. For generalized linear models, dropout performs a form of adaptive regularization. Using this viewpoint, we show that the dropout regularizer is first-order equivalent to an L2 regularizer applied after scaling the features by an estimate of the inverse diagonal Fisher information matrix. We also establish a connection to AdaGrad, an online learning algorithm, and find that a close relative of AdaGrad operates by repeatedly solving linear dropout-regularized problems. By casting dropout as regularization, we develop a natural semi-supervised algorithm that uses unlabeled data to create a better adaptive regularizer. We apply this idea to document classification tasks, and show that it consistently boosts the performance of dropout training, improving on state-of-the-art results on the IMDB reviews dataset.

stat.ML↗