arXiv subjects

Lei Yu

Publications and source records attributed to Lei Yu.

At least 217 records · Page 12Linked to original sources

arXiv2021

The Convexity and Concavity of Envelopes of the Minimum-Relative-Entropy Region for the DSBS

In this paper, we prove that for the doubly symmetric binary distribution, the lower increasing envelope and the upper envelope of the minimum-relative-entropy region are respectively convex and concave. We also prove that another function induced the minimum-relative-entropy region is concave. These two envelopes and this function were previously used to characterize the optimal exponents in strong small-set expansion problems and strong Brascamp--Lieb inequalities. The results in this paper, combined with the strong small-set expansion theorem derived by Yu, Anantharam, and Chen (2021), and the strong Brascamp--Lieb inequality derived by Yu (2021), confirm positively Ordentlich--Polyanskiy--Shayevitz's conjecture on the strong small-set expansion (2019) and Polyanskiy's conjecture on the strong Brascamp--Lieb inequality (2016). The proofs in this paper are based on the equivalence between the convexity of a function and the convexity of the set of minimizers of its Lagrangian dual.

cs.IT↗

arXiv2021

NTIRE 2021 Challenge on Burst Super-Resolution: Methods and Results

This paper reviews the NTIRE2021 challenge on burst super-resolution. Given a RAW noisy burst as input, the task in the challenge was to generate a clean RGB image with 4 times higher resolution. The challenge contained two tracks; Track 1 evaluating on synthetically generated data, and Track 2 using real-world bursts from mobile camera. In the final testing phase, 6 teams submitted results using a diverse set of solutions. The top-performing methods set a new state-of-the-art for the burst super-resolution task.

cs.CV↗

arXiv2021

Diverse Pretrained Context Encodings Improve Document Translation

We propose a new architecture for adapting a sentence-level sequence-to-sequence transformer by incorporating multiple pretrained document context signals and assess the impact on translation performance of (1) different pretraining approaches for generating these signals, (2) the quantity of parallel data for which document context is available, and (3) conditioning on source, target, or source and target contexts. Experiments on the NIST Chinese-English, and IWSLT and WMT English-German tasks support four general conclusions: that using pretrained context representations markedly improves sample efficiency, that adequate parallel data resources are crucial for learning to use document context, that jointly conditioning on multiple context representations outperforms any single representation, and that source context is more valuable for translation performance than target side context. Our best multi-context model consistently outperforms the best existing context-aware transformers.

cs.CL↗

arXiv2021

A Pilot Study of Interplanetary Scintillation with FAST

Observations of Interplanetary Scintillation (IPS) are an efficient remote-sensing method to study the solar wind and inner heliosphere. From 2016 to 2018, some distinctive observations of IPS sources like 3C 286 and 3C 279 were accomplished with the Five-hundred-meter Aperture Spherical radio Telescope (FAST), the largest single-dish telescope in the world. Due to the 270-1620 MHz wide frequency coverage of the Ultra-Wideband (UWB) receiver, one can use both single-frequency and dual-frequency analyses to determine the projected velocity of the solar wind. Moreover, based on the extraordinary sensitivity owing to the large collecting surface area of FAST, we can observe weak IPS signals. With the advantages of both the wider frequency coverage and high sensitivity, also with our radio frequency interference (RFI) mitigation strategy and an optimized model-fitting method developed, in this paper, we analyze the fitting confidence intervals of the solar wind velocity, and present some preliminary results achieved using FAST, which points to the current FAST system being highly capable of carrying out observations of IPS

physics.space-ph↗

arXiv2021

Event-based Synthetic Aperture Imaging with a Hybrid Network

Synthetic aperture imaging (SAI) is able to achieve the see through effect by blurring out the off-focus foreground occlusions and reconstructing the in-focus occluded targets from multi-view images. However, very dense occlusions and extreme lighting conditions may bring significant disturbances to the SAI based on conventional frame-based cameras, leading to performance degeneration. To address these problems, we propose a novel SAI system based on the event camera which can produce asynchronous events with extremely low latency and high dynamic range. Thus, it can eliminate the interference of dense occlusions by measuring with almost continuous views, and simultaneously tackle the over/under exposure problems. To reconstruct the occluded targets, we propose a hybrid encoder-decoder network composed of spiking neural networks (SNNs) and convolutional neural networks (CNNs). In the hybrid network, the spatio-temporal information of the collected events is first encoded by SNN layers, and then transformed to the visual image of the occluded targets by a style-transfer CNN decoder. Through experiments, the proposed method shows remarkable performance in dealing with very dense occlusions and extreme lighting conditions, and high quality visual images can be reconstructed using pure event data.

cs.CV↗

arXiv2021

IsarStep: a Benchmark for High-level Mathematical Reasoning

A well-defined benchmark is essential for measuring and accelerating research progress of machine learning models. In this paper, we present a benchmark for high-level mathematical reasoning and study the reasoning capabilities of neural sequence-to-sequence models. We build a non-synthetic dataset from the largest repository of proofs written by human experts in a theorem prover. The dataset has a broad coverage of undergraduate and research-level mathematical and computer science theorems. In our defined task, a model is required to fill in a missing intermediate proposition given surrounding proofs. This task provides a starting point for the long-term goal of having machines generate human-readable proofs automatically. Our experiments and analysis reveal that while the task is challenging, neural models can capture non-trivial mathematical reasoning. We further design a hierarchical transformer that outperforms the transformer baseline.

cs.LO↗

arXiv2021

Pretraining the Noisy Channel Model for Task-Oriented Dialogue

Direct decoding for task-oriented dialogue is known to suffer from the explaining-away effect, manifested in models that prefer short and generic responses. Here we argue for the use of Bayes' theorem to factorize the dialogue task into two models, the distribution of the context given the response, and the prior for the response itself. This approach, an instantiation of the noisy channel model, both mitigates the explaining-away effect and allows the principled incorporation of large pretrained models for the response prior. We present extensive experiments showing that a noisy channel model decodes better responses compared to direct decoding and that a two stage pretraining strategy, employing both open-domain and task-oriented dialogue data, improves over randomly initialized models.

cs.CL↗

arXiv2021

On Non-Interactive Simulation of Binary Random Variables

We leverage proof techniques Fourier analysis and an existing result in coding theory to derive new bounds for the problem of non-interactive simulation of binary random variables. Previous bounds in the literature were derived by applying data processing inequalities concerning maximal correlation or hypercontractivity. We show that our bounds are sharp in some regimes. For a specific instance of problem parameters, our main result answers an open problem posed by E. Mossel in 2017. As by-products of our analyses, various new properties of the average distance and distance enumerator of binary block codes are established.

cs.IT↗

arXiv2020

Learning a Simple and Effective Model for Multi-turn Response Generation with Auxiliary Tasks

We study multi-turn response generation for open-domain dialogues. The existing state-of-the-art addresses the problem with deep neural architectures. While these models improved response quality, their complexity also hinders the application of the models in real systems. In this work, we pursue a model that has a simple structure yet can effectively leverage conversation contexts for response generation. To this end, we propose four auxiliary tasks including word order recovery, utterance order recovery, masked word recovery, and masked utterance recovery, and optimize the objectives of these tasks together with maximizing the likelihood of generation. By this means, the auxiliary tasks that relate to context understanding can guide the learning of the generation model to achieve a better local optimum. Empirical studies with three benchmarks indicate that our model can significantly outperform state-of-the-art generation models in terms of response quality on both automatic evaluation and human judgment, and at the same time enjoys a much faster decoding process.

cs.CL↗

arXiv2020

ULSD: Unified Line Segment Detection across Pinhole, Fisheye, and Spherical Cameras

Line segment detection is essential for high-level tasks in computer vision and robotics. Currently, most stateof-the-art (SOTA) methods are dedicated to detecting straight line segments in undistorted pinhole images, thus distortions on fisheye or spherical images may largely degenerate their performance. Targeting at the unified line segment detection (ULSD) for both distorted and undistorted images, we propose to represent line segments with the Bezier curve model. Then the line segment detection is tackled by the Bezier curve regression with an end-to-end network, which is model-free and without any undistortion preprocessing. Experimental results on the pinhole, fisheye, and spherical image datasets validate the superiority of the proposed ULSD to the SOTA methods both in accuracy and efficiency (40.6fps for pinhole images). The source code is available at https://github.com/lh9171338/Unified-LineSegment-Detection.

cs.CV↗

arXiv2020

Inferring symmetry in natural language

We present a methodological framework for inferring symmetry of verb predicates in natural language. Empirical work on predicate symmetry has taken two main approaches. The feature-based approach focuses on linguistic features pertaining to symmetry. The context-based approach denies the existence of absolute symmetry but instead argues that such inference is context dependent. We develop methods that formalize these approaches and evaluate them against a novel symmetry inference sentence (SIS) dataset comprised of 400 naturalistic usages of literature-informed verbs spanning the spectrum of symmetry-asymmetry. Our results show that a hybrid transfer learning model that integrates linguistic features with contextualized language models most faithfully predicts the empirical data. Our work integrates existing approaches to symmetry in natural language and suggests how symmetry inference can improve systematicity in state-of-the-art language models.

cs.CL↗

arXiv2020

Improving Spiking Sparse Recovery via Non-Convex Penalties

Compared with digital methods, sparse recovery based on spiking neural networks has great advantages like high computational efficiency and low power-consumption. However, current spiking algorithms cannot guarantee more accurate estimates since they are usually designed to solve the classical optimization with convex penalties, especially the $\ell_{1}$-norm. In fact, convex penalties are observed to underestimate the true solution in practice, while non-convex ones can avoid the underestimation. Inspired by this, we propose an adaptive version of spiking sparse recovery algorithm to solve the non-convex regularized optimization, and provide an analysis on its global asymptotic convergence. Through experiments, the accuracy is greatly improved under different adaptive ways.

eess.SP↗

arXiv2020

Mixed Noise Removal with Pareto Prior

Denoising images contaminated by the mixture of additive white Gaussian noise (AWGN) and impulse noise (IN) is an essential but challenging problem. The presence of impulsive disturbances inevitably affects the distribution of noises and thus largely degrades the performance of traditional AWGN denoisers. Existing methods target to compensate the effects of IN by introducing a weighting matrix, which, however, is lack of proper priori and thus hard to be accurately estimated. To address this problem, we exploit the Pareto distribution as the priori of the weighting matrix, based on which an accurate and robust weight estimator is proposed for mixed noise removal. Particularly, a relatively small portion of pixels are assumed to be contaminated with IN, which should have weights with small values and then be penalized out. This phenomenon can be properly described by the Pareto distribution of type 1. Therefore, armed with the Pareto distribution, we formulate the problem of mixed noise removal in the Bayesian framework, where nonlocal self-similarity priori is further exploited by adopting nonlocal low rank approximation. Compared to existing methods, the proposed method can estimate the weighting matrix adaptively, accurately, and robust for different level of noises, thus can boost the denoising performance. Experimental results on widely used image datasets demonstrate the superiority of our proposed method to the state-of-the-arts.

eess.IV↗

arXiv2020

Structure-Aware Network for Lane Marker Extraction with Dynamic Vision Sensor

Lane marker extraction is a basic yet necessary task for autonomous driving. Although past years have witnessed major advances in lane marker extraction with deep learning models, they all aim at ordinary RGB images generated by frame-based cameras, which limits their performance in extreme cases, like huge illumination change. To tackle this problem, we introduce Dynamic Vision Sensor (DVS), a type of event-based sensor to lane marker extraction task and build a high-resolution DVS dataset for lane marker extraction. We collect the raw event data and generate 5,424 DVS images with a resolution of 1280$\times$800 pixels, the highest one among all DVS datasets available now. All images are annotated with multi-class semantic segmentation format. We then propose a structure-aware network for lane marker extraction in DVS images. It can capture directional information comprehensively with multidirectional slice convolution. We evaluate our proposed network with other state-of-the-art lane marker extraction models on this dataset. Experimental results demonstrate that our method outperforms other competitors. The dataset is made publicly available, including the raw event data, accumulated images and labels.

cs.CV↗

arXiv2020

Event Enhanced High-Quality Image Recovery

With extremely high temporal resolution, event cameras have a large potential for robotics and computer vision. However, their asynchronous imaging mechanism often aggravates the measurement sensitivity to noises and brings a physical burden to increase the image spatial resolution. To recover high-quality intensity images, one should address both denoising and super-resolution problems for event cameras. Since events depict brightness changes, with the enhanced degeneration model by the events, the clear and sharp high-resolution latent images can be recovered from the noisy, blurry and low-resolution intensity observations. Exploiting the framework of sparse learning, the events and the low-resolution intensity observations can be jointly considered. Based on this, we propose an explainable network, an event-enhanced sparse learning network (eSL-Net), to recover the high-quality images from event cameras. After training with a synthetic dataset, the proposed eSL-Net can largely improve the performance of the state-of-the-art by 7-12 dB. Furthermore, without additional training process, the proposed eSL-Net can be easily extended to generate continuous frames with frame-rate as high as the events.

cs.CV↗

arXiv2020

Implicit Euler ODE Networks for Single-Image Dehazing

Deep convolutional neural networks (CNN) have been applied for image dehazing tasks, where the residual network (ResNet) is often adopted as the basic component to avoid the vanishing gradient problem. Recently, many works indicate that the ResNet can be considered as the explicit Euler forward approximation of an ordinary differential equation (ODE). In this paper, we extend the explicit forward approximation to the implicit backward counterpart, which can be realized via a recursive neural network, named IM-block. Given that, we propose an efficient end-to-end multi-level implicit network (MI-Net) for the single image dehazing problem. Moreover, multi-level fusing (MLF) mechanism and residual channel attention block (RCA-block) are adopted to boost performance of our network. Experiments on several dehazing benchmark datasets demonstrate that our method outperforms existing methods and achieves the state-of-the-art performance.

cs.CV↗

arXiv2020

Better Document-Level Machine Translation with Bayes' Rule

We show that Bayes' rule provides an effective mechanism for creating document translation models that can be learned from only parallel sentences and monolingual documents---a compelling benefit as parallel documents are not always available. In our formulation, the posterior probability of a candidate translation is the product of the unconditional (prior) probability of the candidate output document and the "reverse translation probability" of translating the candidate output back into the source language. Our proposed model uses a powerful autoregressive language model as the prior on target language documents, but it assumes that each sentence is translated independently from the target to the source language. Crucially, at test time, when a source document is observed, the document language model prior induces dependencies between the translations of the source sentences in the posterior. The model's independence assumption not only enables efficient use of available data, but it additionally admits a practical left-to-right beam-search algorithm for carrying out inference. Experiments show that our model benefits from using cross-sentence context in the language model, and it outperforms existing document translation approaches.

cs.CL↗

arXiv2020

Single Image Deraining with Continuous Rain Density Estimation

Single image deraining (SIDR) often suffers from over/under deraining due to the nonuniformity of rain densities and the variety of raindrop scales. In this paper, we propose a \textbf{\it co}ntinuous \textbf{\it de}nsity guided network (CODE-Net) for SIDR. Particularly, it is composed of { a rain {\color{black}streak} extractor and a denoiser}, where the convolutional sparse coding (CSC) is exploited to filter out noises from the extracted rain streaks. Inspired by the reweighted iterative soft-threshold for CSC, we address the problem of continuous rain density estimation by learning the weights with channel attention blocks from sparse codes. We further {\color{black}develop} a multiscale strategy to depict rain streaks appearing at different scales. Experiments on synthetic and real-world data demonstrate the superiority of our methods over recent {\color{black}state of the arts}, in terms of both quantitative and qualitative results. Additionally, instead of quantizing rain density with several levels, our CODE-Net can provide continuous-valued estimations of rain densities, which is more desirable in real applications.

eess.IV↗