arXiv subjects

Kevin Lin

Publications and source records attributed to Kevin Lin.

106 records · Page 6Linked to original sources

arXiv2020

Learning Nonparametric Human Mesh Reconstruction from a Single Image without Ground Truth Meshes

Nonparametric approaches have shown promising results on reconstructing 3D human mesh from a single monocular image. Unlike previous approaches that use a parametric human model like skinned multi-person linear model (SMPL), and attempt to regress the model parameters, nonparametric approaches relax the heavy reliance on the parametric space. However, existing nonparametric methods require ground truth meshes as their regression target for each vertex, and obtaining ground truth mesh labels is very expensive. In this paper, we propose a novel approach to learn human mesh reconstruction without any ground truth meshes. This is made possible by introducing two new terms into the loss function of a graph convolutional neural network (Graph CNN). The first term is the Laplacian prior that acts as a regularizer on the reconstructed mesh. The second term is the part segmentation loss that forces the projected region of the reconstructed mesh to match the part segmentation. Experimental results on multiple public datasets show that without using 3D ground truth meshes, the proposed approach outperforms the previous state-of-the-art approaches that require ground truth meshes for training.

cs.CV↗

arXiv2020

Learning to Generate Multiple Style Transfer Outputs for an Input Sentence

Text style transfer refers to the task of rephrasing a given text in a different style. While various methods have been proposed to advance the state of the art, they often assume the transfer output follows a delta distribution, and thus their models cannot generate different style transfer results for a given input text. To address the limitation, we propose a one-to-many text style transfer framework. In contrast to prior works that learn a one-to-one mapping that converts an input sentence to one output sentence, our approach learns a one-to-many mapping that can convert an input sentence to multiple different output sentences, while preserving the input content. This is achieved by applying adversarial training with a latent decomposition scheme. Specifically, we decompose the latent representation of the input sentence to a style code that captures the language style variation and a content code that encodes the language style-independent content. We then combine the content code with the style code for generating a style transfer output. By combining the same content code with a different style code, we generate a different style transfer output. Extensive experimental results with comparisons to several text style transfer approaches on multiple public datasets using a diverse set of performance metrics validate effectiveness of the proposed approach.

cs.CL↗

arXiv2020

Neural Module Networks for Reasoning over Text

Answering compositional questions that require multiple steps of reasoning against text is challenging, especially when they involve discrete, symbolic operations. Neural module networks (NMNs) learn to parse such questions as executable programs composed of learnable modules, performing well on synthetic visual QA domains. However, we find that it is challenging to learn these models for non-synthetic questions on open-domain text, where a model needs to deal with the diversity of natural language and perform a broader range of reasoning. We extend NMNs by: (a) introducing modules that reason over a paragraph of text, performing symbolic reasoning (such as arithmetic, sorting, counting) over numbers and dates in a probabilistic and differentiable manner; and (b) proposing an unsupervised auxiliary loss to help extract arguments associated with the events in text. Additionally, we show that a limited amount of heuristically-obtained question program and intermediate module output supervision provides sufficient inductive bias for accurate learning. Our proposed model significantly outperforms state-of-the-art models on a subset of the DROP dataset that poses a variety of reasoning challenges that are covered by our modules.

cs.CL↗

arXiv2019

Reasoning Over Paragraph Effects in Situations

A key component of successfully reading a passage of text is the ability to apply knowledge gained from the passage to a new situation. In order to facilitate progress on this kind of reading, we present ROPES, a challenging benchmark for reading comprehension targeting Reasoning Over Paragraph Effects in Situations. We target expository language describing causes and effects (e.g., "animal pollinators increase efficiency of fertilization in flowers"), as they have clear implications for new situations. A system is presented a background passage containing at least one of these relations, a novel situation that uses this background, and questions that require reasoning about effects of the relationships in the background passage in the context of the situation. We collect background passages from science textbooks and Wikipedia that contain such phenomena, and ask crowd workers to author situations, questions, and answers, resulting in a 14,322 question dataset. We analyze the challenges of this task and evaluate the performance of state-of-the-art reading comprehension models. The best model performs only slightly better than randomly guessing an answer of the correct type, at 61.6% F1, well below the human performance of 89.0%.

cs.CL↗

arXiv2019

QuaRTz: An Open-Domain Dataset of Qualitative Relationship Questions

We introduce the first open-domain dataset, called QuaRTz, for reasoning about textual qualitative relationships. QuaRTz contains general qualitative statements, e.g., "A sunscreen with a higher SPF protects the skin longer.", twinned with 3864 crowdsourced situated questions, e.g., "Billy is wearing sunscreen with a lower SPF than Lucy. Who will be best protected from the sun?", plus annotations of the properties being compared. Unlike previous datasets, the general knowledge is textual and not tied to a fixed set of relationships, and tests a system's ability to comprehend and apply textual qualitative knowledge in a novel setting. We find state-of-the-art results are substantially (20%) below human performance, presenting an open challenge to the NLP community.

cs.CL↗

arXiv2019

Grammar-based Neural Text-to-SQL Generation

The sequence-to-sequence paradigm employed by neural text-to-SQL models typically performs token-level decoding and does not consider generating SQL hierarchically from a grammar. Grammar-based decoding has shown significant improvements for other semantic parsing tasks, but SQL and other general programming languages have complexities not present in logical formalisms that make writing hierarchical grammars difficult. We introduce techniques to handle these complexities, showing how to construct a schema-dependent grammar with minimal over-generation. We analyze these techniques on ATIS and Spider, two challenging text-to-SQL datasets, demonstrating that they yield 14--18\% relative reductions in error.

cs.CL↗

arXiv2019

Can the Stochastic Wave Equation with Strong Drift Hit Zero?

We study the stochastic wave equation with multiplicative noise and singular drift: \[ \partial_tu(t,x)=Δu(t,x)+u^{-α}(t,x)+g(u(t,x))\dot{W}(t,x) \] where $x$ lies in the circle $\mathbf{R}/J\mathbf{Z}$ and $u(0,x)>0$. We show that (i) If $0<α<1$ then with positive probability, $u(t,x)=0$ for some $(t,x)$. (ii) If $α>3$ then with probability one, $u(t,x)\ne0$ for all $(t,x)$.

math.PR↗

arXiv2019

DeepBase: Deep Inspection of Neural Networks

Although deep learning models perform remarkably well across a range of tasks such as language translation and object recognition, it remains unclear what high-level logic, if any, they follow. Understanding this logic may lead to more transparency, better model design, and faster experimentation. Recent machine learning research has leveraged statistical methods to identify hidden units that behave (e.g., activate) similarly to human understandable logic, but those analyses require considerable manual effort. Our insight is that many of those studies follow a common analysis pattern, which we term Deep Neural Inspection. There is opportunity to provide a declarative abstraction to easily express, execute, and optimize them. This paper describes DeepBase, a system to inspect neural network behaviors through a unified interface. We model logic with user-provided hypothesis functions that annotate the data with high-level labels (e.g., part-of-speech tags, image captions). DeepBase lets users quickly identify individual or groups of units that have strong statistical dependencies with desired hypotheses. We discuss how DeepBase can express existing analyses, propose a set of simple and effective optimizations to speed up a standard Python implementation by up to 72x, and reproduce recent studies from the NLP literature.

cs.DB↗

arXiv2019

Subgoals, Problem Solving Phases, and Sources of Knowledge: A Complex Mangle

Educational researchers have increasingly drawn attention to how students develop computational thinking (CT) skills, including in science, math, and literacy contexts. A key component of CT is the process of abstraction, a particularly challenging concept for novice programmers, but one vital to problem solving. We propose a framework based on situated cognition that can be used to document how instructors and students communicate about abstractions during the problem solving process. We develop this framework in a multimodal interaction analysis of a 32-minute long excerpt of a middle school student working in the PixelBots JavaScript programming environment at a two-week summer programming workshop taught by undergraduate CS majors. Through a microgenetic analysis of the process of teaching and learning about abstraction in this excerpt, we document the extemporaneous prioritization of subgoals and the back-and-forth coordination of problem solving phases. In our case study, we identify that (a) problem solving phases are nested with several instances of context-switching within a single phase; (b) the introduction of new ideas and information create bridges or opportunities to move between different problem solving phases; (c) planning to solve a problem is a non-linear process; and (d) pedagogical moves such as modeling and prompting highlight situated resources and advance problem solving. Future research should address how to help students structure subgoals and reflect on connections between problem solving phases, and how to help instructors reflect on their routes to supporting students in the problem solving process.

cs.CY↗

arXiv2018

Post-Selection Inference for Changepoint Detection Algorithms with Application to Copy Number Variation Data

Changepoint detection methods are used in many areas of science and engineering, e.g., in the analysis of copy number variation data, to detect abnormalities in copy numbers along the genome. Despite the broad array of available tools, methodology for quantifying our uncertainty in the strength (or presence) of given changepoints, post-detection, are lacking. Post-selection inference offers a framework to fill this gap, but the most straightforward application of these methods results in low-powered tests and leaves open several important questions about practical usability. In this work, we carefully tailor post-selection inference methods towards changepoint detection, focusing as our main scientific application on copy number variation data. As for changepoint algorithms, we study binary segmentation, and two of its most popular variants, wild and circular, and the fused lasso. We implement some of the latest developments in post-selection inference theory: we use auxiliary randomization to improve power, which requires implementations of MCMC algorithms (importance sampling and hit-and-run sampling) to carry out our tests. We also provide recommendations for improving practical useability, detailed simulations, and an example analysis on array comparative genomic hybridization (CGH) data.

stat.ME↗

arXiv2018

Adversarial Learning for Fine-grained Image Search

Fine-grained image search is still a challenging problem due to the difficulty in capturing subtle differences regardless of pose variations of objects from fine-grained categories. In practice, a dynamic inventory with new fine-grained categories adds another dimension to this challenge. In this work, we propose an end-to-end network, called FGGAN, that learns discriminative representations by implicitly learning a geometric transformation from multi-view images for fine-grained image search. We integrate a generative adversarial network (GAN) that can automatically handle complex view and pose variations by converting them to a canonical view without any predefined transformations. Moreover, in an open-set scenario, our network is able to better match images from unseen and unknown fine-grained categories. Extensive experiments on two public datasets and a newly collected dataset have demonstrated the outstanding robust performance of the proposed FGGAN in both closed-set and open-set scenarios, providing as much as 10% relative improvement compared to baselines.

cs.CV↗

arXiv2018

Adversarial Ranking for Language Generation

Generative adversarial networks (GANs) have great successes on synthesizing data. However, the existing GANs restrict the discriminator to be a binary classifier, and thus limit their learning capacity for tasks that need to synthesize output with rich structures such as natural language descriptions. In this paper, we propose a novel generative adversarial network, RankGAN, for generating high-quality language descriptions. Rather than training the discriminator to learn and assign absolute binary predicate for individual data sample, the proposed RankGAN is able to analyze and rank a collection of human-written and machine-written sentences by giving a reference group. By viewing a set of data samples collectively and evaluating their quality through relative ranking scores, the discriminator is able to make better assessment which in turn helps to learn a better generator. The proposed RankGAN is optimized through the policy gradient technique. Experimental results on multiple public datasets clearly demonstrate the effectiveness of the proposed approach.

cs.CL↗

arXiv2017

Supervised Learning of Semantics-Preserving Hash via Deep Convolutional Neural Networks

This paper presents a simple yet effective supervised deep hash approach that constructs binary hash codes from labeled data for large-scale image search. We assume that the semantic labels are governed by several latent attributes with each attribute on or off, and classification relies on these attributes. Based on this assumption, our approach, dubbed supervised semantics-preserving deep hashing (SSDH), constructs hash functions as a latent layer in a deep network and the binary codes are learned by minimizing an objective function defined over classification error and other desirable hash codes properties. With this design, SSDH has a nice characteristic that classification and retrieval are unified in a single learning model. Moreover, SSDH performs joint learning of image representations, hash codes, and classification in a point-wised manner, and thus is scalable to large-scale datasets. SSDH is simple and can be realized by a slight enhancement of an existing deep architecture for classification; yet it is effective and outperforms other hashing approaches on several benchmarks and large datasets. Compared with state-of-the-art approaches, SSDH achieves higher retrieval accuracy, while the classification performance is not sacrificed.

cs.CV↗

arXiv2016

Approximate Recovery in Changepoint Problems, from $\ell_2$ Estimation Error Rates

In the 1-dimensional multiple changepoint detection problem, we prove that any procedure with a fast enough $\ell_2$ error rate, in terms of its estimation of the underlying piecewise constant mean vector, automatically has an (approximate) changepoint screening property---specifically, each true jump in the underlying mean vector has an estimated jump nearby. We also show, again assuming only knowledge of the $\ell_2$ error rate, that a simple post-processing step can be used to eliminate spurious estimated changepoints, and thus delivers an (approximate) changepoint recovery property---specifically, in addition to the screening property described above, we are assured that each estimated jump has a true jump nearby. As a special case, we focus on the application of these results to the 1-dimensional fused lasso, i.e., 1-dimensional total variation denoising, and compare the implications with existing results from the literature. We also study extensions to related problems, such as changepoint detection over graphs.

stat.ME↗

arXiv2016

Data-based stochastic model reduction for the Kuramoto--Sivashinsky equation

The problem of constructing data-based, predictive, reduced models for the Kuramoto-Sivashinsky equation is considered, under circumstances where one has observation data only for a small subset of the dynamical variables. Accurate prediction is achieved by developing a discrete-time stochastic reduced system, based on a NARMAX (Nonlinear Autoregressive Moving Average with eXogenous input) representation. The practical issue, with the NARMAX representation as with any other, is to identify an efficient structure, i.e., one with a small number of terms and coefficients. This is accomplished here by estimating coefficients for an approximate inertial form. The broader significance of the results is discussed.

math.NA↗

arXiv2013

Optimization for Compressed Sensing: the Simplex Method and Kronecker Sparsification

In this paper we present two new approaches to efficiently solve large-scale compressed sensing problems. These two ideas are independent of each other and can therefore be used either separately or together. We consider all possibilities. For the first approach, we note that the zero vector can be taken as the initial basic (infeasible) solution for the linear programming problem and therefore, if the true signal is very sparse, some variants of the simplex method can be expected to take only a small number of pivots to arrive at a solution. We implemented one such variant and demonstrate a dramatic improvement in computation time on very sparse signals. The second approach requires a redesigned sensing mechanism in which the vector signal is stacked into a matrix. This allows us to exploit the Kronecker compressed sensing (KCS) mechanism. We show that the Kronecker sensing requires stronger conditions for perfect recovery compared to the original vector problem. However, the Kronecker sensing, modeled correctly, is a much sparser linear optimization problem. Hence, algorithms that benefit from sparse problem representation, such as interior-point methods, can solve the Kronecker sensing problems much faster than the corresponding vector problem. In our numerical studies, we demonstrate a ten-fold improvement in the computation time.

stat.ML↗