arXiv subjects

Hao Zhou

Publications and source records attributed to Hao Zhou.

At least 433 records · Page 24Linked to original sources

arXiv2021

Pre-training Multilingual Neural Machine Translation by Leveraging Alignment Information

We investigate the following question for machine translation (MT): can we develop a single universal MT model to serve as the common seed and obtain derivative and improved models on arbitrary language pairs? We propose mRASP, an approach to pre-train a universal multilingual neural machine translation model. Our key idea in mRASP is its novel technique of random aligned substitution, which brings words and phrases with similar meanings across multiple languages closer in the representation space. We pre-train a mRASP model on 32 language pairs jointly with only public datasets. The model is then fine-tuned on downstream language pairs to obtain specialized MT models. We carry out extensive experiments on 42 translation directions across a diverse settings, including low, medium, rich resource, and as well as transferring to exotic language pairs. Experimental results demonstrate that mRASP achieves significant performance improvement compared to directly training on those target pairs. It is the first time to verify that multiple low-resource language pairs can be utilized to improve rich resource MT. Surprisingly, mRASP is even able to improve the translation quality on exotic languages that never occur in the pre-training corpus. Code, data, and pre-trained models are available at https://github.com/linzehui/mRASP.

cs.CL↗

arXiv2020

An Analysis of Scatter Characteristics in X-ray CT Spectral Correction

X-ray scatter remains a major physics challenge in volumetric computed tomography (CT), whose physical and statistical behaviors have been commonly leveraged in order to eliminate its impact on CT image quality. In this work, we conduct an in-depth derivation of how the scatter distribution and scatter to primary ratio (SPR) will change during the spectral correction, leading to an interesting finding on the property of scatter: when applying the spectral correction before scatter is removed, the impact of SPR on a CT projection will be scaled by the first derivative of the mapping function; while the scatter distribution in the transmission domain will be scaled by the product of the first derivative of the mapping function and a natural exponential of the projection difference before and after the mapping. Such a characterization of scatter's behavior provides an analytic approach of compensating for the SPR as well as approximating the change of scatter distribution after spectral correction, even though both of them might be significantly distorted as the linearization mapping function in spectral correction could vary a lot from one detector pixel to another. We conduct an evaluation of SPR compensations on a Catphan phantom and an anthropomorphic chest phantom to validate the characteristics of scatter. In addition, this scatter property is also directly adopted into CT imaging using a spectral modulator with flying focal spot technology (SMFFS) as an example to demonstrate its potential in practical applications.

physics.med-ph↗

arXiv2020

Online Distributed Job Dispatching with Outdated and Partially-Observable Information

In this paper, we investigate online distributed job dispatching in an edge computing system residing in a Metropolitan Area Network (MAN). Specifically, job dispatchers are implemented on access points (APs) which collect jobs from mobile users and distribute each job to a server at the edge or the cloud. A signaling mechanism with periodic broadcast is introduced to facilitate cooperation among APs. The transmission latency is non-negligible in MAN, which leads to outdated information sharing among APs. Moreover, the fully-observed system state is discouraged as reception of all broadcast is time consuming. Therefore, we formulate the distributed optimization of job dispatching strategies among the APs as a Markov decision process with partial and outdated system state, i.e., partially observable Markov Decision Process (POMDP). The conventional solution for POMDP is impractical due to huge time complexity. We propose a novel low-complexity solution framework for distributed job dispatching, based on which the optimization of job dispatching policy can be decoupled via an alternative policy iteration algorithm, so that the distributed policy iteration of each AP can be made according to partial and outdated observation. A theoretical performance lower bound is proved for our approximate MDP solution. Furthermore, we conduct extensive simulations based on the Google Cluster trace. The evaluation results show that our policy can achieve as high as $20.67\%$ reduction in average job response time compared with heuristic baselines, and our algorithm consistently performs well under various parameter settings.

cs.NI↗

arXiv2020

Comprehensive Graph-conditional Similarity Preserving Network for Unsupervised Cross-modal Hashing

Unsupervised cross-modal hashing (UCMH) has become a hot topic recently. Current UCMH focuses on exploring data similarities. However, current UCMH methods calculate the similarity between two data, mainly relying on the two data's cross-modal features. These methods suffer from inaccurate similarity problems that result in a suboptimal retrieval Hamming space, because the cross-modal features between the data are not sufficient to describe the complex data relationships, such as situations where two data have different feature representations but share the inherent concepts. In this paper, we devise a deep graph-neighbor coherence preserving network (DGCPN). Specifically, DGCPN stems from graph models and explores graph-neighbor coherence by consolidating the information between data and their neighbors. DGCPN regulates comprehensive similarity preserving losses by exploiting three types of data similarities (i.e., the graph-neighbor coherence, the coexistent similarity, and the intra- and inter-modality consistency) and designs a half-real and half-binary optimization strategy to reduce the quantization errors during hashing. Essentially, DGCPN addresses the inaccurate similarity problem by exploring and exploiting the data's intrinsic relationships in a graph. We conduct extensive experiments on three public UCMH datasets. The experimental results demonstrate the superiority of DGCPN, e.g., by improving the mean average precision from 0.722 to 0.751 on MIRFlickr-25K using 64-bit hashing codes to retrieve texts from images. We will release the source code package and the trained model on https://github.com/Atmegal/DGCPN.

cs.IR↗

arXiv2020

Where, What, Whether: Multi-modal Learning Meets Pedestrian Detection

Pedestrian detection benefits greatly from deep convolutional neural networks (CNNs). However, it is inherently hard for CNNs to handle situations in the presence of occlusion and scale variation. In this paper, we propose W$^3$Net, which attempts to address above challenges by decomposing the pedestrian detection task into \textbf{\textit{W}}here, \textbf{\textit{W}}hat and \textbf{\textit{W}}hether problem directing against pedestrian localization, scale prediction and classification correspondingly. Specifically, for a pedestrian instance, we formulate its feature by three steps. i) We generate a bird view map, which is naturally free from occlusion issues, and scan all points on it to look for suitable locations for each pedestrian instance. ii) Instead of utilizing pre-fixed anchors, we model the interdependency between depth and scale aiming at generating depth-guided scales at different locations for better matching instances of different sizes. iii) We learn a latent vector shared by both visual and corpus space, by which false positives with similar vertical structure but lacking human partial features would be filtered out. We achieve state-of-the-art results on widely used datasets (Citypersons and Caltech). In particular. when evaluating on heavy occlusion subset, our results reduce MR$^{-2}$ from 49.3$\%$ to 18.7$\%$ on Citypersons, and from 45.18$\%$ to 28.33$\%$ on Caltech.

cs.CV↗

arXiv2020

Reciprocal Supervised Learning Improves Neural Machine Translation

Despite the recent success on image classification, self-training has only achieved limited gains on structured prediction tasks such as neural machine translation (NMT). This is mainly due to the compositionality of the target space, where the far-away prediction hypotheses lead to the notorious reinforced mistake problem. In this paper, we revisit the utilization of multiple diverse models and present a simple yet effective approach named Reciprocal-Supervised Learning (RSL). RSL first exploits individual models to generate pseudo parallel data, and then cooperatively trains each model on the combined synthetic corpus. RSL leverages the fact that different parameterized models have different inductive biases, and better predictions can be made by jointly exploiting the agreement among each other. Unlike the previous knowledge distillation methods built upon a much stronger teacher, RSL is capable of boosting the accuracy of one model by introducing other comparable or even weaker models. RSL can also be viewed as a more efficient alternative to ensemble. Extensive experiments demonstrate the superior performance of RSL on several benchmarks with significant margins.

cs.CL↗

arXiv2020

CPM: A Large-scale Generative Chinese Pre-trained Language Model

Pre-trained Language Models (PLMs) have proven to be beneficial for various downstream NLP tasks. Recently, GPT-3, with 175 billion parameters and 570GB training data, drew a lot of attention due to the capacity of few-shot (even zero-shot) learning. However, applying GPT-3 to address Chinese NLP tasks is still challenging, as the training corpus of GPT-3 is primarily English, and the parameters are not publicly available. In this technical report, we release the Chinese Pre-trained Language Model (CPM) with generative pre-training on large-scale Chinese training data. To the best of our knowledge, CPM, with 2.6 billion parameters and 100GB Chinese training data, is the largest Chinese pre-trained language model, which could facilitate several downstream Chinese NLP tasks, such as conversation, essay generation, cloze test, and language understanding. Extensive experiments demonstrate that CPM achieves strong performance on many NLP tasks in the settings of few-shot (even zero-shot) learning. The code and parameters are available at https://github.com/TsinghuaAI/CPM-Generate.

cs.CL↗

arXiv2020

On the Sentence Embeddings from Pre-trained Language Models

Pre-trained contextual representations like BERT have achieved great success in natural language processing. However, the sentence embeddings from the pre-trained language models without fine-tuning have been found to poorly capture semantic meaning of sentences. In this paper, we argue that the semantic information in the BERT embeddings is not fully exploited. We first reveal the theoretical connection between the masked language model pre-training objective and the semantic similarity task theoretically, and then analyze the BERT sentence embeddings empirically. We find that BERT always induces a non-smooth anisotropic semantic space of sentences, which harms its performance of semantic similarity. To address this issue, we propose to transform the anisotropic sentence embedding distribution to a smooth and isotropic Gaussian distribution through normalizing flows that are learned with an unsupervised objective. Experimental results show that our proposed BERT-flow method obtains significant performance gains over the state-of-the-art sentence embeddings on a variety of semantic textual similarity tasks. The code is available at https://github.com/bohanli/BERT-flow.

cs.CL↗

arXiv2020

Active Sentence Learning by Adversarial Uncertainty Sampling in Discrete Space

Active learning for sentence understanding aims at discovering informative unlabeled data for annotation and therefore reducing the demand for labeled data. We argue that the typical uncertainty sampling method for active learning is time-consuming and can hardly work in real-time, which may lead to ineffective sample selection. We propose adversarial uncertainty sampling in discrete space (AUSDS) to retrieve informative unlabeled samples more efficiently. AUSDS maps sentences into latent space generated by the popular pre-trained language models, and discover informative unlabeled text samples for annotation via adversarial attack. The proposed approach is extremely efficient compared with traditional uncertainty sampling with more than 10x speedup. Experimental results on five datasets show that AUSDS outperforms strong baselines on effectiveness.

cs.CL↗

arXiv2020

Accelerated computational micromechanics

We present an approach to solving problems in micromechanics that is amenable to massively parallel calculations through the use of graphical processing units and other accelerators. The problems lead to nonlinear differential equations that are typically second order in space and first order in time. This combination of nonlinearity and nonlocality makes such problems difficult to solve in parallel. However, this combination is a result of collapsing nonlocal, but linear and universal physical laws (kinematic compatibility, balance laws), and nonlinear but local constitutive relations. We propose an operator-splitting scheme inspired by this structure. The governing equations are formulated as (incremental) variational problems, the differential constraints like compatibility are introduced using an augmented Lagrangian, and the resulting incremental variational principle is solved by the alternating direction method of multipliers. The resulting algorithm has a natural connection to physical principles, and also enables massively parallel implementation on structured grids. We present this method and use it to study two examples. The first concerns the long wavelength instability of finite elasticity, and allows us to verify the approach against previous numerical simulations. We also use this example to study convergence and parallel performance. The second example concerns microstructure evolution in liquid crystal elastomers and provides new insights into some counter-intuitive properties of these materials. We use this example to validate the model and the approach against experimental observations.

cs.CE↗

arXiv2020

WeChat Neural Machine Translation Systems for WMT20

We participate in the WMT 2020 shared news translation task on Chinese to English. Our system is based on the Transformer (Vaswani et al., 2017a) with effective variants and the DTMT (Meng and Zhang, 2019) architecture. In our experiments, we employ data selection, several synthetic data generation approaches (i.e., back-translation, knowledge distillation, and iterative in-domain knowledge transfer), advanced finetuning approaches and self-bleu based model ensemble. Our constrained Chinese to English system achieves 36.9 case-sensitive BLEU score, which is the highest among all submissions.

cs.CL↗

arXiv2020

Dispersed Exponential Family Mixture VAEs for Interpretable Text Generation

Deep generative models are commonly used for generating images and text. Interpretability of these models is one important pursuit, other than the generation quality. Variational auto-encoder (VAE) with Gaussian distribution as prior has been successfully applied in text generation, but it is hard to interpret the meaning of the latent variable. To enhance the controllability and interpretability, one can replace the Gaussian prior with a mixture of Gaussian distributions (GM-VAE), whose mixture components could be related to hidden semantic aspects of data. In this paper, we generalize the practice and introduce DEM-VAE, a class of models for text generation using VAEs with a mixture distribution of exponential family. Unfortunately, a standard variational training algorithm fails due to the mode-collapse problem. We theoretically identify the root cause of the problem and propose an effective algorithm to train DEM-VAE. Our method penalizes the training with an extra dispersion term to induce a well-structured latent space. Experimental results show that our approach does obtain a meaningful space, and it outperforms strong baselines in text generation benchmarks. The code is available at https://github.com/wenxianxian/demvae.

cs.LG↗

arXiv2020

An Empirical Evaluation of GDPR Compliance Violations in Android mHealth Apps

The purpose of the General Data Protection Regulation (GDPR) is to provide improved privacy protection. If an app controls personal data from users, it needs to be compliant with GDPR. However, GDPR lists general rules rather than exact step-by-step guidelines about how to develop an app that fulfills the requirements. Therefore, there may exist GDPR compliance violations in existing apps, which would pose severe privacy threats to app users. In this paper, we take mobile health applications (mHealth apps) as a peephole to examine the status quo of GDPR compliance in Android apps. We first propose an automated system, named \mytool, to bridge the semantic gap between the general rules of GDPR and the app implementations by identifying the data practices declared in the app privacy policy and the data relevant behaviors in the app code. Then, based on \mytool, we detect three kinds of GDPR compliance violations, including the incompleteness of privacy policy, the inconsistency of data collections, and the insecurity of data transmission. We perform an empirical evaluation of 796 mHealth apps. The results reveal that 189 (23.7\%) of them do not provide complete privacy policies. Moreover, 59 apps collect sensitive data through different measures, but 46 (77.9\%) of them contain at least one inconsistent collection behavior. Even worse, among the 59 apps, only 8 apps try to ensure the transmission security of collected data. However, all of them contain at least one encryption or SSL misuse. Our work exposes severe privacy issues to raise awareness of privacy protection for app users and developers.

cs.SE↗

arXiv2020

Do You Have the Right Scissors? Tailoring Pre-trained Language Models via Monte-Carlo Methods

It has been a common approach to pre-train a language model on a large corpus and fine-tune it on task-specific data. In practice, we observe that fine-tuning a pre-trained model on a small dataset may lead to over- and/or under-estimation problem. In this paper, we propose MC-Tailor, a novel method to alleviate the above issue in text generation tasks by truncating and transferring the probability mass from over-estimated regions to under-estimated ones. Experiments on a variety of text generation datasets show that MC-Tailor consistently and significantly outperforms the fine-tuning approach. Our code is available at this url.

cs.CL↗

arXiv2020

Generating Fluent Adversarial Examples for Natural Languages

Efficiently building an adversarial attacker for natural language processing (NLP) tasks is a real challenge. Firstly, as the sentence space is discrete, it is difficult to make small perturbations along the direction of gradients. Secondly, the fluency of the generated examples cannot be guaranteed. In this paper, we propose MHA, which addresses both problems by performing Metropolis-Hastings sampling, whose proposal is designed with the guidance of gradients. Experiments on IMDB and SNLI show that our proposed MHA outperforms the baseline model on attacking capability. Adversarial training with MAH also leads to better robustness and performance.

cs.CL↗

arXiv2020

Improving Maximum Likelihood Training for Text Generation with Density Ratio Estimation

Auto-regressive sequence generative models trained by Maximum Likelihood Estimation suffer the exposure bias problem in practical finite sample scenarios. The crux is that the number of training samples for Maximum Likelihood Estimation is usually limited and the input data distributions are different at training and inference stages. Many method shave been proposed to solve the above problem (Yu et al., 2017; Lu et al., 2018), which relies on sampling from the non-stationary model distribution and suffers from high variance or biased estimations. In this paper, we proposeψ-MLE, a new training scheme for auto-regressive sequence generative models, which is effective and stable when operating at large sample space encountered in text generation. We derive our algorithm from a new perspective of self-augmentation and introduce bias correction with density ratio estimation. Extensive experimental results on synthetic data and real-world text generation tasks demonstrate that our method stably outperforms Maximum Likelihood Estimation and other state-of-the-art sequence generative models in terms of both quality and diversity.

stat.ML↗

arXiv2020

Xiaomingbot: A Multilingual Robot News Reporter

This paper proposes the building of Xiaomingbot, an intelligent, multilingual and multimodal software robot equipped with four integral capabilities: news generation, news translation, news reading and avatar animation. Its system summarizes Chinese news that it automatically generates from data tables. Next, it translates the summary or the full article into multiple languages, and reads the multilingual rendition through synthesized speech. Notably, Xiaomingbot utilizes a voice cloning technology to synthesize the speech trained from a real person's voice data in one input language. The proposed system enjoys several merits: it has an animated avatar, and is able to generate and read multilingual news. Since it was put into practice, Xiaomingbot has written over 600,000 articles, and gained over 150,000 followers on social media platforms.

eess.AS↗

arXiv2020

COVID-19 causes record decline in global CO2 emissions

The considerable cessation of human activities during the COVID-19 pandemic has affected global energy use and CO2 emissions. Here we show the unprecedented decrease in global fossil CO2 emissions from January to April 2020 was of 7.8% (938 Mt CO2 with a +6.8% of 2-σ uncertainty) when compared with the period last year. In addition other emerging estimates of COVID impacts based on monthly energy supply or estimated parameters, this study contributes to another step that constructed the near-real-time daily CO2 emission inventories based on activity from power generation (for 29 countries), industry (for 73 countries), road transportation (for 406 cities), aviation and maritime transportation and commercial and residential sectors emissions (for 206 countries). The estimates distinguished the decline of CO2 due to COVID-19 from the daily, weekly and seasonal variations as well as the holiday events. The COVID-related decreases in CO2 emissions in road transportation (340.4 Mt CO2, -15.5%), power (292.5 Mt CO2, -6.4% compared to 2019), industry (136.2 Mt CO2, -4.4%), aviation (92.8 Mt CO2, -28.9%), residential (43.4 Mt CO2, -2.7%), and international shipping (35.9Mt CO2, -15%). Regionally, decreases in China were the largest and earliest (234.5 Mt CO2,-6.9%), followed by Europe (EU-27 & UK) (138.3 Mt CO2, -12.0%) and the U.S. (162.4 Mt CO2, -9.5%). The declines of CO2 are consistent with regional nitrogen oxides concentrations observed by satellites and ground-based networks, but the calculated signal of emissions decreases (about 1Gt CO2) will have little impacts (less than 0.13ppm by April 30, 2020) on the overserved global CO2 concertation. However, with observed fast CO2 recovery in China and partial re-opening globally, our findings suggest the longer-term effects on CO2 emissions are unknown and should be carefully monitored using multiple measures.

econ.GN↗