skip to main content

Title: Latent Part-of-Speech Sequences for Neural Machine Translation
Learning target side syntactic structure has been shown to improve Neural Machine Translation (NMT). However, incorporating syntax through latent variables introduces additional complexity in inference, as the models need to marginalize over the latent syntactic structures. To avoid this, models often resort to greedy search which only allows them to explore a limited portion of the latent space. In this work, we introduce a new latent variable model, LaSyn, that captures the co-dependence between syntax and semantics, while allowing for effective and efficient inference over the latent space. LaSyn decouples direct dependence between successive latent variables, which allows its decoder to exhaustively search through the latent syntactic choices, while keeping decoding speed proportional to the size of the latent variable vocabulary. We implement LaSyn by modifying a transformer-based NMT system and design a neural expectation maximization algorithm that we regularize with part-of-speech information as the latent sequences. Evaluations on four different MT tasks show that incorporating target side syntax with LaSyn improves both translation quality, and also provides an opportunity to improve diversity.
; ; ; ;
Award ID(s):
Publication Date:
Journal Name:
Empirical Methods in Natural Language Processing
Page Range or eLocation-ID:
780 to 790
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract Scientific and engineering problems often require the use of artificial intelligence to aid understanding and the search for promising designs. While Gaussian processes (GP) stand out as easy-to-use and interpretable learners, they have difficulties in accommodating big data sets, categorical inputs, and multiple responses, which has become a common challenge for a growing number of data-driven design applications. In this paper, we propose a GP model that utilizes latent variables and functions obtained through variational inference to address the aforementioned challenges simultaneously. The method is built upon the latent-variable Gaussian process (LVGP) model where categorical factors are mapped into a continuous latent space to enable GP modeling of mixed-variable data sets. By extending variational inference to LVGP models, the large training data set is replaced by a small set of inducing points to address the scalability issue. Output response vectors are represented by a linear combination of independent latent functions, forming a flexible kernel structure to handle multiple responses that might have distinct behaviors. Comparative studies demonstrate that the proposed method scales well for large data sets with over 104 data points, while outperforming state-of-the-art machine learning methods without requiring much hyperparameter tuning. In addition, an interpretable latent spacemore »is obtained to draw insights into the effect of categorical factors, such as those associated with “building blocks” of architectures and element choices in metamaterial and materials design. Our approach is demonstrated for machine learning of ternary oxide materials and topology optimization of a multiscale compliant mechanism with aperiodic microstructures and multiple materials.« less
  2. Noisy channel models have been especially effective in neural machine translation (NMT). However, recent approaches like "beam search and rerank" (BSR) incur significant computation overhead during inference, making real-world application infeasible. We aim to study if it is possible to build an amortized noisy channel NMT model such that when we do greedy decoding during inference, the translation accuracy matches that of BSR in terms of reward (based on the source-to-target log probability and the target-to-source log probability) and quality (based on BLEU and BLEURT). We attempt three approaches to train the new model: knowledge distillation, one-step-deviation imitation learning, and Q learning. The first approach obtains the noisy channel signal from a pseudo-corpus, and the latter two approaches aim to optimize toward a noisy-channel MT reward directly. For all three approaches, the generated translations fail to achieve rewards comparable to BSR, but the translation quality approximated by BLEU and BLEURT is similar to the quality of BSR-produced translations. Additionally, all three approaches speed up inference by 1-2 orders of magnitude.
  3. The link between mind, brain, and behavior has mystified philosophers and scientists for millennia. Recent progress has been made by forming statistical associations between manifest variables of the brain (e.g., electroencephalogram [EEG], functional MRI [fMRI]) and manifest variables of behavior (e.g., response times, accuracy) through hierarchical latent variable models. Within this framework, one can make inferences about the mind in a statistically principled way, such that complex patterns of brain–behavior associations drive the inference procedure. However, previous approaches were limited in the flexibility of the linking function, which has proved prohibitive for understanding the complex dynamics exhibited by the brain. In this article, we propose a data-driven, nonparametric approach that allows complex linking functions to emerge from fitting a hierarchical latent representation of the mind to multivariate, multimodal data. Furthermore, to enforce biological plausibility, we impose both spatial and temporal structure so that the types of realizable system dynamics are constrained. To illustrate the benefits of our approach, we investigate the model’s performance in a simulation study and apply it to experimental data. In the simulation study, we verify that the model can be accurately fitted to simulated data, and latent dynamics can be well recovered. In an experimental application,more »we simultaneously fit the model to fMRI and behavioral data from a continuous motion tracking task. We show that the model accurately recovers both neural and behavioral data and reveals interesting latent cognitive dynamics, the topology of which can be contrasted with several aspects of the experiment.

    « less
  4. Motivated by modern applications in which one constructs graphical models based on a very large number of features, this paper introduces a new class of cluster-based graphical models, in which variable clustering is applied as an initial step for reducing the dimension of the feature space. We employ model assisted clustering, in which the clusters contain features that are similar to the same unobserved latent variable. Two different cluster-based Gaussian graphical models are considered: the latent variable graph, corresponding to the graphical model associated with the unobserved latent variables, and the cluster-average graph, corresponding to the vector of features averaged over clusters. Our study reveals that likelihood based inference for the latent graph, not analyzed previously, is analytically intractable. Our main contribution is the development and analysis of alternative estimation and inference strategies, for the precision matrix of an unobservable latent vector Z. We replace the likelihood of the data by an appropriate class of empirical risk functions, that can be specialized to the latent graphical model and to the simpler, but under-analyzed, cluster-average graphical model. The estimators thus derived can be used for inference on the graph structure, for instance on edge strength or pattern recovery. Inference is basedmore »on the asymptotic limits of the entry-wise estimates of the precision matrices associated with the conditional independence graphs under consideration. While taking the uncertainty induced by the clustering step into account, we establish Berry-Esseen central limit theorems for the proposed estimators. It is noteworthy that, although the clusters are estimated adaptively from the data, the central limit theorems regarding the entries of the estimated graphs are proved under the same conditions one would use if the clusters were known in advance. As an illustration of the usage of these newly developed inferential tools, we show that they can be reliably used for recovery of the sparsity pattern of the graphs we study, under FDR control, which is verified via simulation studies and an fMRI data analysis. These experimental results confirm the theoretically established difference between the two graph structures. Furthermore, the data analysis suggests that the latent variable graph, corresponding to the unobserved cluster centers, can help provide more insight into the understanding of the brain connectivity networks relative to the simpler, average-based, graph.« less
  5. Modern neural network architectures use structured linear transformations, such as low-rank matrices, sparse matrices, permutations, and the Fourier transform, to improve inference speed and reduce memory usage compared to general linear maps. However, choosing which of the myriad structured transformations to use (and its associated parameterization) is a laborious task that requires trading off speed, space, and accuracy. We consider a different approach: we introduce a family of matrices called kaleidoscope matrices (K-matrices) that provably capture any structured matrix with near-optimal space (parameter) and time (arithmetic operation) complexity. We empirically validate that K-matrices can be automatically learned within end-to-end pipelines to replace hand-crafted procedures, in order to improve model quality. For example, replacing channel shuffles in ShuffleNet improves classification accuracy on ImageNet by up to 5%. K-matrices can also simplify hand-engineered pipelines---we replace filter bank feature computation in speech data preprocessing with a learnable kaleidoscope layer, resulting in only 0.4% loss in accuracy on the TIMIT speech recognition task. In addition, K-matrices can capture latent structure in models: for a challenging permuted image classification task, adding a K-matrix to a standard convolutional architecture can enable learning the latent permutation and improve accuracy by over 8 points. We provide a practicallymore »efficient implementation of our approach, and use K-matrices in a Transformer network to attain 36% faster end-to-end inference speed on a language translation task.« less