arXiv subjects

Min Yang

Publications and source records attributed to Min Yang.

At least 289 records · Page 16Linked to original sources

arXiv2022

GALAXY: A Generative Pre-trained Model for Task-Oriented Dialog with Semi-Supervised Learning and Explicit Policy Injection

Pre-trained models have proved to be powerful in enhancing task-oriented dialog systems. However, current pre-training methods mainly focus on enhancing dialog understanding and generation tasks while neglecting the exploitation of dialog policy. In this paper, we propose GALAXY, a novel pre-trained dialog model that explicitly learns dialog policy from limited labeled dialogs and large-scale unlabeled dialog corpora via semi-supervised learning. Specifically, we introduce a dialog act prediction task for policy optimization during pre-training and employ a consistency regularization term to refine the learned representation with the help of unlabeled dialogs. We also implement a gating mechanism to weigh suitable unlabeled dialog samples. Empirical results show that GALAXY substantially improves the performance of task-oriented dialog systems, and achieves new state-of-the-art results on benchmark datasets: In-Car, MultiWOZ2.0 and MultiWOZ2.1, improving their end-to-end combined scores by 2.5, 5.3 and 5.5 points, respectively. We also show that GALAXY has a stronger few-shot ability than existing models under various low-resource settings.

cs.CL↗

arXiv2022

Linking-Enhanced Pre-Training for Table Semantic Parsing

Recently pre-training models have significantly improved the performance of various NLP tasks by leveraging large-scale text corpora to improve the contextual representation ability of the neural network. The large pre-training language model has also been applied in the area of table semantic parsing. However, existing pre-training approaches have not carefully explored explicit interaction relationships between a question and the corresponding database schema, which is a key ingredient for uncovering their semantic and structural correspondence. Furthermore, the question-aware representation learning in the schema grounding context has received less attention in pre-training objective.To alleviate these issues, this paper designs two novel pre-training objectives to impose the desired inductive bias into the learned representations for table pre-training. We further propose a schema-aware curriculum learning approach to mitigate the impact of noise and learn effectively from the pre-training data in an easy-to-hard manner. We evaluate our pre-trained framework by fine-tuning it on two benchmarks, Spider and SQUALL. The results demonstrate the effectiveness of our pre-training objective and curriculum compared to a variety of baselines.

cs.CL↗

arXiv2022

Improve Deep Image Inpainting by Emphasizing the Complexity of Missing Regions

Deep image inpainting research mainly focuses on constructing various neural network architectures or imposing novel optimization objectives. However, on the one hand, building a state-of-the-art deep inpainting model is an extremely complex task, and on the other hand, the resulting performance gains are sometimes very limited. We believe that besides the frameworks of inpainting models, lightweight traditional image processing techniques, which are often overlooked, can actually be helpful to these deep models. In this paper, we enhance the deep image inpainting models with the help of classical image complexity metrics. A knowledge-assisted index composed of missingness complexity and forward loss is presented to guide the batch selection in the training procedure. This index helps find samples that are more conducive to optimization in each iteration and ultimately boost the overall inpainting performance. The proposed approach is simple and can be plugged into many deep inpainting models by changing only a few lines of code. We experimentally demonstrate the improvements for several recently developed image inpainting models on various datasets.

cs.CV↗

arXiv2021

Exploring the Security Boundary of Data Reconstruction via Neuron Exclusivity Analysis

Among existing privacy attacks on the gradient of neural networks, \emph{data reconstruction attack}, which reverse engineers the training batch from the gradient, poses a severe threat on the private training data. Despite its empirical success on large architectures and small training batches, unstable reconstruction accuracy is also observed when a smaller architecture or a larger batch is under attack. Due to the weak interpretability of existing learning-based attacks, there is little known on why, when and how data reconstruction attack is feasible. In our work, we perform the first analytic study on the security boundary of data reconstruction from gradient via a microcosmic view on neural networks with rectified linear units (ReLUs), the most popular activation function in practice. For the first time, we characterize the insecure/secure boundary of data reconstruction attack in terms of the \emph{neuron exclusivity state} of a training batch, indexed by the number of \emph{\textbf{Ex}clusively \textbf{A}ctivated \textbf{N}eurons} (ExANs, i.e., a ReLU activated by only one sample in a batch). Intuitively, we show a training batch with more ExANs are more vulnerable to data reconstruction attack and vice versa. On the one hand, we construct a novel deterministic attack algorithm which substantially outperforms previous attacks for reconstructing training batches lying in the insecure boundary of a neural network. Meanwhile, for training batches lying in the secure boundary, we prove the impossibility of unique reconstruction, based on which an exclusivity reduction strategy is devised to enlarge the secure boundary for mitigation purposes.

cs.CR↗

arXiv2021

Enhancing Top-N Item Recommendations by Peer Collaboration

Deep neural networks (DNN) have achieved great success in the recommender systems (RS) domain. However, to achieve remarkable performance, DNN-based recommender models often require numerous parameters, which inevitably bring redundant neurons and weights, a phenomenon referred to as over-parameterization. In this paper, we plan to exploit such redundancy phenomena to improve the performance of RS. Specifically, we propose PCRec, a top-N item \underline{rec}ommendation framework that leverages collaborative training of two DNN-based recommender models with the same network structure, termed \underline{p}eer \underline{c}ollaboration. PCRec can reactivate and strengthen the unimportant (redundant) weights during training, which achieves higher prediction accuracy but maintains its original inference efficiency. To realize this, we first introduce two criteria to identify the importance of weights of a given recommender model. Then, we rejuvenate the unimportant weights by transplanting outside information (i.e., weights) from its peer network. After such an operation and retraining, the original recommender model is endowed with more representation capacity by possessing more functional model parameters. To show its generality, we instantiate PCRec by using three well-known recommender models. We conduct extensive experiments on three real-world datasets, and show that PCRec yields significantly better recommendations than its counterpart with the same model (parameter) size.

cs.IR↗

arXiv2021

A Chinese Multi-type Complex Questions Answering Dataset over Wikidata

Complex Knowledge Base Question Answering is a popular area of research in the past decade. Recent public datasets have led to encouraging results in this field, but are mostly limited to English and only involve a small number of question types and relations, hindering research in more realistic settings and in languages other than English. In addition, few state-of-the-art KBQA models are trained on Wikidata, one of the most popular real-world knowledge bases. We propose CLC-QuAD, the first large scale complex Chinese semantic parsing dataset over Wikidata to address these challenges. Together with the dataset, we present a text-to-SPARQL baseline model, which can effectively answer multi-type complex questions, such as factual questions, dual intent questions, boolean questions, and counting questions, with Wikidata as the background knowledge. We finally analyze the performance of SOTA KBQA models on this dataset and identify the challenges facing Chinese KBQA.

cs.CL↗

arXiv2021

Relaxation and darkening of excitonic complexes in electrostatically-doped monolayer semiconductors: Roles of exciton-electron and trion-electron interactions

We present photoluminescence measurements in monolayer WSe$_2$, which point to the importance of the interaction between charged particles and excitonic complexes. The theoretical analysis highlights the key role played by exchange scattering, referring to cases wherein the particle composition of the complex changes after the interaction. For example, exchange scattering renders bright excitonic complexes dark in monolayer WSe$_2$ on accounts of the unique valley-spin configuration in this material. In addition to the ultrafast energy relaxation of hot excitonic complexes following their interaction with electrons or holes, our analysis sheds light on several key features that are commonly seen in the photoluminescence of this monolayer semiconductor. In particular, we can understand why the photoluminescence intensity of the neutral bright exciton is strongest when the monolayer is hole-doped rather than charge neutral or electron-doped. Or the reason for the dramatic increase of the photoluminescence intensity of negatively-charged excitons (trions) as soon as electrons are added to the monolayer. To self-consistently explain the findings, we further study the photoluminescence spectra at different excitation energies and analyze the behavior of the elusive indirect exciton.

cond-mat.mtrl-sci↗

arXiv2021

An Orthogonal Classifier for Improving the Adversarial Robustness of Neural Networks

Neural networks are susceptible to artificially designed adversarial perturbations. Recent efforts have shown that imposing certain modifications on classification layer can improve the robustness of the neural networks. In this paper, we explicitly construct a dense orthogonal weight matrix whose entries have the same magnitude, thereby leading to a novel robust classifier. The proposed classifier avoids the undesired structural redundancy issue in previous work. Applying this classifier in standard training on clean data is sufficient to ensure the high accuracy and good robustness of the model. Moreover, when extra adversarial samples are used, better robustness can be further obtained with the help of a special worst-case loss. Experimental results show that our method is efficient and competitive to many state-of-the-art defensive approaches. Our code is available at \url{https://github.com/MTandHJ/roboc}.

cs.CV↗

arXiv2021

A Low Complexity MAP Detector for OTFS Modulation in Logarithmic Domain

Orthogonal time-frequency space (OTFS) has been confirmed to take advantage of full time-frequency diversity to significantly improve error performance in high-mobility scenarios. We found that the proposed message passing (MP) and variational Bayes (VB) detectors can achieve approximate maximum a posteriori (MAP) detection, the interferences cannot be completely eliminated in the absence of noise. To achieve near-optimal MAP detection, this letter proposes a novel detection method based on sum-product algorithm (SPA) with low complexity. Leveraging subtly factorized posteriori probabilities, the obtained pairwise interactions can effectively avoid enumeration of high-dimensional variables, thereby making it applicable to fractional Doppler cases. We further simplify the proposed algorithm in the logarithmic domain so that the message propagation processing only involves addition. Finally, simulations results demonstrate the superior error performance gains of our proposed algorithm at high signal-to-noise ratios (SNRs).

cs.IT↗

arXiv2021

DOLG: Single-Stage Image Retrieval with Deep Orthogonal Fusion of Local and Global Features

Image Retrieval is a fundamental task of obtaining images similar to the query one from a database. A common image retrieval practice is to firstly retrieve candidate images via similarity search using global image features and then re-rank the candidates by leveraging their local features. Previous learning-based studies mainly focus on either global or local image representation learning to tackle the retrieval task. In this paper, we abandon the two-stage paradigm and seek to design an effective single-stage solution by integrating local and global information inside images into compact image representations. Specifically, we propose a Deep Orthogonal Local and Global (DOLG) information fusion framework for end-to-end image retrieval. It attentively extracts representative local information with multi-atrous convolutions and self-attention at first. Components orthogonal to the global image representation are then extracted from the local information. At last, the orthogonal components are concatenated with the global representation as a complementary, and then aggregation is performed to generate the final representation. The whole framework is end-to-end differentiable and can be trained with image-level labels. Extensive experimental results validate the effectiveness of our solution and show that our model achieves state-of-the-art image retrieval performances on Revisited Oxford and Paris datasets.

cs.CV↗

arXiv2021

Continual Learning for Task-oriented Dialogue System with Iterative Network Pruning, Expanding and Masking

This ability to learn consecutive tasks without forgetting how to perform previously trained problems is essential for developing an online dialogue system. This paper proposes an effective continual learning for the task-oriented dialogue system with iterative network pruning, expanding and masking (TPEM), which preserves performance on previously encountered tasks while accelerating learning progress on subsequent tasks. Specifically, TPEM (i) leverages network pruning to keep the knowledge for old tasks, (ii) adopts network expanding to create free weights for new tasks, and (iii) introduces task-specific network masking to alleviate the negative impact of fixed weights of old tasks on new tasks. We conduct extensive experiments on seven different tasks from three benchmark datasets and show empirically that TPEM leads to significantly improved results over the strong competitors. For reproducibility, we submit the code and data at: https://github.com/siat-nlp/TPEM

cs.CL↗

arXiv2021

Iterative Network Pruning with Uncertainty Regularization for Lifelong Sentiment Classification

Lifelong learning capabilities are crucial for sentiment classifiers to process continuous streams of opinioned information on the Web. However, performing lifelong learning is non-trivial for deep neural networks as continually training of incrementally available information inevitably results in catastrophic forgetting or interference. In this paper, we propose a novel iterative network pruning with uncertainty regularization method for lifelong sentiment classification (IPRLS), which leverages the principles of network pruning and weight regularization. By performing network pruning with uncertainty regularization in an iterative manner, IPRLS can adapta single BERT model to work with continuously arriving data from multiple domains while avoiding catastrophic forgetting and interference. Specifically, we leverage an iterative pruning method to remove redundant parameters in large deep networks so that the freed-up space can then be employed to learn new tasks, tackling the catastrophic forgetting problem. Instead of keeping the old-tasks fixed when learning new tasks, we also use an uncertainty regularization based on the Bayesian online learning framework to constrain the update of old tasks weights in BERT, which enables positive backward transfer, i.e. learning new tasks improves performance on past tasks while protecting old knowledge from being lost. In addition, we propose a task-specific low-dimensional residual function in parallel to each layer of BERT, which makes IPRLS less prone to losing the knowledge saved in the base BERT network when learning a new task. Extensive experiments on 16 popular review corpora demonstrate that the proposed IPRLS method sig-nificantly outperforms the strong baselines for lifelong sentiment classification. For reproducibility, we submit the code and data at:https://github.com/siat-nlp/IPRLS.

cs.CL↗

arXiv2021

User-specific Adaptive Fine-tuning for Cross-domain Recommendations

Making accurate recommendations for cold-start users has been a longstanding and critical challenge for recommender systems (RS). Cross-domain recommendations (CDR) offer a solution to tackle such a cold-start problem when there is no sufficient data for the users who have rarely used the system. An effective approach in CDR is to leverage the knowledge (e.g., user representations) learned from a related but different domain and transfer it to the target domain. Fine-tuning works as an effective transfer learning technique for this objective, which adapts the parameters of a pre-trained model from the source domain to the target domain. However, current methods are mainly based on the global fine-tuning strategy: the decision of which layers of the pre-trained model to freeze or fine-tune is taken for all users in the target domain. In this paper, we argue that users in RS are personalized and should have their own fine-tuning policies for better preference transfer learning. As such, we propose a novel User-specific Adaptive Fine-tuning method (UAF), selecting which layers of the pre-trained network to fine-tune, on a per-user basis. Specifically, we devise a policy network with three alternative strategies to automatically decide which layers to be fine-tuned and which layers to have their parameters frozen for each user. Extensive experiments show that the proposed UAF exhibits significantly better and more robust performance for user cold-start recommendation.

cs.IR↗

arXiv2021

A Template-guided Hybrid Pointer Network for Knowledge-basedTask-oriented Dialogue Systems

Most existing neural network based task-oriented dialogue systems follow encoder-decoder paradigm, where the decoder purely depends on the source texts to generate a sequence of words, usually suffering from instability and poor readability. Inspired by the traditional template-based generation approaches, we propose a template-guided hybrid pointer network for the knowledge-based task-oriented dialogue system, which retrieves several potentially relevant answers from a pre-constructed domain-specific conversational repository as guidance answers, and incorporates the guidance answers into both the encoding and decoding processes. Specifically, we design a memory pointer network model with a gating mechanism to fully exploit the semantic correlation between the retrieved answers and the ground-truth response. We evaluate our model on four widely used task-oriented datasets, including one simulated and three manually created datasets. The experimental results demonstrate that the proposed model achieves significantly better performance than the state-of-the-art methods over different automatic evaluation metrics.

cs.CL↗

arXiv2021

Applications of Triggered Scaler Module for Accelerator Timing

Since 2006, there have been unexpected trigger-failure events during the operation of the Japan Proton Accelerator Research Complex (J-PARC) timing system. However, among the many suspected modules, it has been difficult to find the one with the fault. To find such a faulty module more easily, the triggered scaler module, which was designed as a PLC-type I/O module, was developed. This type of module requires the start signal of the J-PARC main ring (MR) slow cycle (2.48/5.20 s) and the start signal of the rapid cycle (25 Hz) as reference signals, which are generated by the J-PARC timing system. A scaler in the module counts the number of input pulses during a machine cycle, and stores them in memory-buffers. In 2018, the module was evaluated using the trigger for injection kicker and the low-level radio frequency (LLRF) signal of J-PARC MR. The expected results were shown. In 2020, two applications were developed based on a triggered scaler module. The first one, an application for detecting unexpected-trigger events, successfully detected simulated trigger-failure events. The second one, an application for the abort signal of a machine protection system (MPS), showed that the module is capable of visualizing signals based on a relationship with the accelerator cycle. Two applications show possibilities for the development of more customized applications in the future. An idea to develop a portable unexpected-trigger detection system, which can be used in other accelerator facilities, is discussed herein.

physics.acc-ph↗

arXiv2021

StackRec: Efficient Training of Very Deep Sequential Recommender Models by Iterative Stacking

Deep learning has brought great progress for the sequential recommendation (SR) tasks. With advanced network architectures, sequential recommender models can be stacked with many hidden layers, e.g., up to 100 layers on real-world recommendation datasets. Training such a deep network is difficult because it can be computationally very expensive and takes much longer time, especially in situations where there are tens of billions of user-item interactions. To deal with such a challenge, we present StackRec, a simple, yet very effective and efficient training framework for deep SR models by iterative layer stacking. Specifically, we first offer an important insight that hidden layers/blocks in a well-trained deep SR model have very similar distributions. Enlightened by this, we propose the stacking operation on the pre-trained layers/blocks to transfer knowledge from a shallower model to a deep model, then we perform iterative stacking so as to yield a much deeper but easier-to-train SR model. We validate the performance of StackRec by instantiating it with four state-of-the-art SR models in three practical scenarios with real-world datasets. Extensive experiments show that StackRec achieves not only comparable performance, but also substantial acceleration in training time, compared to SR models that are trained from scratch. Codes are available at https://github.com/wangjiachun0426/StackRec.

cs.IR↗

arXiv2021

Spin/Valley pumping of resident electrons in WSe2 and WS2 monolayers

Monolayers of transition metal dichalcogenides are ideal materials to control both spin and valley degrees of freedom either electrically or optically. Nevertheless, optical excitation mostly generates excitons species with inherently short lifetime and spin/valley relaxation time. Here we demonstrate a very efficient spin/valley optical pumping of resident electrons in n-doped WSe2 and WS2 monolayers. We observe that, using a continuous wave laser and appropriate doping and excitation densities, negative trion doublet lines exhibit circular polarization of opposite sign and the photoluminescence intensity of the triplet trion is more than four times larger with circular excitation than with linear excitation. We interpret our results as a consequence of a large dynamic polarization of resident electrons using circular light.

cond-mat.mes-hall↗

arXiv2021

Contextualized Knowledge-aware Attentive Neural Network: Enhancing Answer Selection with Knowledge

Answer selection, which is involved in many natural language processing applications such as dialog systems and question answering (QA), is an important yet challenging task in practice, since conventional methods typically suffer from the issues of ignoring diverse real-world background knowledge. In this paper, we extensively investigate approaches to enhancing the answer selection model with external knowledge from knowledge graph (KG). First, we present a context-knowledge interaction learning framework, Knowledge-aware Neural Network (KNN), which learns the QA sentence representations by considering a tight interaction with the external knowledge from KG and the textual information. Then, we develop two kinds of knowledge-aware attention mechanism to summarize both the context-based and knowledge-based interactions between questions and answers. To handle the diversity and complexity of KG information, we further propose a Contextualized Knowledge-aware Attentive Neural Network (CKANN), which improves the knowledge representation learning with structure information via a customized Graph Convolutional Network (GCN) and comprehensively learns context-based and knowledge-based sentence representation via the multi-view knowledge-aware attention mechanism. We evaluate our method on four widely-used benchmark QA datasets, including WikiQA, TREC QA, InsuranceQA and Yahoo QA. Results verify the benefits of incorporating external knowledge from KG, and show the robust superiority and extensive applicability of our method.

cs.CL↗