arXiv subjects

Zhen Yang

Publications and source records attributed to Zhen Yang.

At least 181 records · Page 10Linked to original sources

arXiv2023

Deep Reinforcement Learning-driven Cross-Community Energy Interaction Optimal Scheduling

In order to coordinate energy interactions among various communities and energy conversions among multi-energy subsystems within the multi-community integrated energy system under uncertain conditions, and achieve overall optimization and scheduling of the comprehensive energy system, this paper proposes a comprehensive scheduling model that utilizes a multi-agent deep reinforcement learning algorithm to learn load characteristics of different communities and make decisions based on this knowledge. In this model, the scheduling problem of the integrated energy system is transformed into a Markov decision process and solved using a data-driven deep reinforcement learning algorithm, which avoids the need for modeling complex energy coupling relationships between multi-communities and multi-energy subsystems. The simulation results show that the proposed method effectively captures the load characteristics of different communities and utilizes their complementary features to coordinate reasonable energy interactions among them. This leads to a reduction in wind curtailment rate from 16.3% to 0% and lowers the overall operating cost by 5445.6 Yuan, demonstrating significant economic and environmental benefits.

eess.SY↗

arXiv2023

ViLTA: Enhancing Vision-Language Pre-training through Textual Augmentation

Vision-language pre-training (VLP) methods are blossoming recently, and its crucial goal is to jointly learn visual and textual features via a transformer-based architecture, demonstrating promising improvements on a variety of vision-language tasks. Prior arts usually focus on how to align visual and textual features, but strategies for improving the robustness of model and speeding up model convergence are left insufficiently explored. In this paper, we propose a novel method ViLTA, comprising of two components to further facilitate the model to learn fine-grained representations among image-text pairs. For Masked Language Modeling (MLM), we propose a cross-distillation method to generate soft labels to enhance the robustness of model, which alleviates the problem of treating synonyms of masked words as negative samples in one-hot labels. For Image-Text Matching (ITM), we leverage the current language encoder to synthesize hard negatives based on the context of language input, encouraging the model to learn high-quality representations by increasing the difficulty of the ITM task. By leveraging the above techniques, our ViLTA can achieve better performance on various vision-language tasks. Extensive experiments on benchmark datasets demonstrate that the effectiveness of ViLTA and its promising potential for vision-language pre-training.

cs.CV↗

arXiv2023

BatchSampler: Sampling Mini-Batches for Contrastive Learning in Vision, Language, and Graphs

In-Batch contrastive learning is a state-of-the-art self-supervised method that brings semantically-similar instances close while pushing dissimilar instances apart within a mini-batch. Its key to success is the negative sharing strategy, in which every instance serves as a negative for the others within the mini-batch. Recent studies aim to improve performance by sampling hard negatives \textit{within the current mini-batch}, whose quality is bounded by the mini-batch itself. In this work, we propose to improve contrastive learning by sampling mini-batches from the input data. We present BatchSampler\footnote{The code is available at \url{https://github.com/THUDM/BatchSampler}} to sample mini-batches of hard-to-distinguish (i.e., hard and true negatives to each other) instances. To make each mini-batch have fewer false negatives, we design the proximity graph of randomly-selected instances. To form the mini-batch, we leverage random walk with restart on the proximity graph to help sample hard-to-distinguish instances. BatchSampler is a simple and general technique that can be directly plugged into existing contrastive learning models in vision, language, and graphs. Extensive experiments on datasets of three modalities show that BatchSampler can consistently improve the performance of powerful contrastive models, as shown by significant improvements of SimCLR on ImageNet-100, SimCSE on STS (language), and GraphCL and MVGRL on graph datasets.

cs.LG↗

arXiv2023

Mitigating Catastrophic Forgetting in Task-Incremental Continual Learning with Adaptive Classification Criterion

Task-incremental continual learning refers to continually training a model in a sequence of tasks while overcoming the problem of catastrophic forgetting (CF). The issue arrives for the reason that the learned representations are forgotten for learning new tasks, and the decision boundary is destructed. Previous studies mostly consider how to recover the representations of learned tasks. It is seldom considered to adapt the decision boundary for new representations and in this paper we propose a Supervised Contrastive learning framework with adaptive classification criterion for Continual Learning (SCCL), In our method, a contrastive loss is used to directly learn representations for different tasks and a limited number of data samples are saved as the classification criterion. During inference, the saved data samples are fed into the current model to obtain updated representations, and a k Nearest Neighbour module is used for classification. In this way, the extensible model can solve the learned tasks with adaptive criteria of saved samples. To mitigate CF, we further use an instance-wise relation distillation regularization term and a memory replay module to maintain the information of previous tasks. Experiments show that SCCL achieves state-of-the-art performance and has a stronger ability to overcome CF compared with the classification baselines.

cs.LG↗

arXiv2023

Investigating Forgetting in Pre-Trained Representations Through Continual Learning

Representation forgetting refers to the drift of contextualized representations during continual training. Intuitively, the representation forgetting can influence the general knowledge stored in pre-trained language models (LMs), but the concrete effect is still unclear. In this paper, we study the effect of representation forgetting on the generality of pre-trained language models, i.e. the potential capability for tackling future downstream tasks. Specifically, we design three metrics, including overall generality destruction (GD), syntactic knowledge forgetting (SynF), and semantic knowledge forgetting (SemF), to measure the evolution of general knowledge in continual learning. With extensive experiments, we find that the generality is destructed in various pre-trained LMs, and syntactic and semantic knowledge is forgotten through continual learning. Based on our experiments and analysis, we further get two insights into alleviating general knowledge forgetting: 1) training on general linguistic tasks at first can mitigate general knowledge forgetting; 2) the hybrid continual learning method can mitigate the generality destruction and maintain more general knowledge compared with those only considering rehearsal or regularization.

cs.CL↗

arXiv2023

Causal Interventions-based Few-Shot Named Entity Recognition

Few-shot named entity recognition (NER) systems aims at recognizing new classes of entities based on a few labeled samples. A significant challenge in the few-shot regime is prone to overfitting than the tasks with abundant samples. The heavy overfitting in few-shot learning is mainly led by spurious correlation caused by the few samples selection bias. To alleviate the problem of the spurious correlation in the few-shot NER, in this paper, we propose a causal intervention-based few-shot NER method. Based on the prototypical network, the method intervenes in the context and prototype via backdoor adjustment during training. In particular, intervening in the context of the one-shot scenario is very difficult, so we intervene in the prototype via incremental learning, which can also avoid catastrophic forgetting. Our experiments on different benchmarks show that our approach achieves new state-of-the-art results (achieving up to 29% absolute improvement and 12% on average for all tasks).

cs.CL↗

arXiv2023

Programming hydrogel adhesion with engineered polymer network topology

Hydrogel adhesion that can be easily modulated in magnitude, space, and time is desirable in many emerging applications ranging from tissue engineering, and soft robotics, to wearable devices. In synthetic materials, these complex adhesion behaviors are often achieved individually with mechanisms and apparatus that are difficult to integrate. Here, we report a universal strategy to embody multifaceted adhesion programmability in synthetic hydrogels. By designing the surface network topology of a hydrogel, supramolecular linkages that result in contrasting adhesion behaviors are formed on the hydrogel interface. The incorporation of different topological linkages leads to dynamically tunable adhesion with high-resolution spatial programmability without alteration of bulk mechanics and chemistry. Further, the association of linkages enables stable and tunable adhesion kinetics that can be tailored to suit different applications. We rationalize the physics of chain slippage, rupture, and diffusion that underpins emergent programmable behaviors. We then incorporate the strategy into the designs of various devices such as smart wound patches, fluidic channels, drug-eluting devices, and reconfigurable soft robotics. Our study presents a simple and robust platform in which adhesion controllability in multiple aspects can be easily integrated into a single design of a hydrogel network.

cond-mat.soft↗

arXiv2023

CLIP$^2$: Contrastive Language-Image-Point Pretraining from Real-World Point Cloud Data

Contrastive Language-Image Pre-training, benefiting from large-scale unlabeled text-image pairs, has demonstrated great performance in open-world vision understanding tasks. However, due to the limited Text-3D data pairs, adapting the success of 2D Vision-Language Models (VLM) to the 3D space remains an open problem. Existing works that leverage VLM for 3D understanding generally resort to constructing intermediate 2D representations for the 3D data, but at the cost of losing 3D geometry information. To take a step toward open-world 3D vision understanding, we propose Contrastive Language-Image-Point Cloud Pretraining (CLIP$^2$) to directly learn the transferable 3D point cloud representation in realistic scenarios with a novel proxy alignment mechanism. Specifically, we exploit naturally-existed correspondences in 2D and 3D scenarios, and build well-aligned and instance-based text-image-point proxies from those complex scenarios. On top of that, we propose a cross-modal contrastive objective to learn semantic and instance-level aligned point cloud representation. Experimental results on both indoor and outdoor scenarios show that our learned 3D representation has great transfer ability in downstream tasks, including zero-shot and few-shot 3D recognition, which boosts the state-of-the-art methods by large margins. Furthermore, we provide analyses of the capability of different representations in real scenarios and present the optional ensemble scheme.

cs.CV↗

arXiv2023

The Graph feature fusion technique for speaker recognition based on wav2vec2.0 framework

Pre-trained wav2vec2.0 model has been proved its effectiveness for speaker recognition. However, current feature processing methods are focusing on classical pooling on the output features of the pre-trained wav2vec2.0 model, such as mean pooling, max pooling etc. That methods take the features as the independent and irrelevant units, ignoring the inter-relationship among all the features, and do not take the features as an overall representation of a speaker. Gated Recurrent Unit (GRU), as a feature fusion method, can also be considered as a complicated pooling technique, mainly focuses on the temporal information, which may show poor performance in some situations that the main information is not on the temporal dimension. In this paper, we investigate the graph neural network (GNN) as a backend processing module based on wav2vec2.0 framework to provide a solution for the mentioned matters. The GNN takes all the output features as the graph signal data and extracts the related graph structure information of features for speaker recognition. Specifically, we first give a simple proof that the GNN feature fusion method can outperform than the mean, max, random pooling methods and so on theoretically. Then, we model the output features of wav2vec2.0 as the vertices of a graph, and construct the graph adjacency matrix by graph attention network (GAT). Finally, we follow the message passing neural network (MPNN) to design our message function, vertex update function and readout function to transform the speaker features into the graph features. The experiments show our performance can provide a relative improvement compared to the baseline methods. Code is available at xxx.

eess.AS↗

arXiv2023

Towards Domain Generalization for Multi-view 3D Object Detection in Bird-Eye-View

Multi-view 3D object detection (MV3D-Det) in Bird-Eye-View (BEV) has drawn extensive attention due to its low cost and high efficiency. Although new algorithms for camera-only 3D object detection have been continuously proposed, most of them may risk drastic performance degradation when the domain of input images differs from that of training. In this paper, we first analyze the causes of the domain gap for the MV3D-Det task. Based on the covariate shift assumption, we find that the gap mainly attributes to the feature distribution of BEV, which is determined by the quality of both depth estimation and 2D image's feature representation. To acquire a robust depth prediction, we propose to decouple the depth estimation from the intrinsic parameters of the camera (i.e. the focal length) through converting the prediction of metric depth to that of scale-invariant depth and perform dynamic perspective augmentation to increase the diversity of the extrinsic parameters (i.e. the camera poses) by utilizing homography. Moreover, we modify the focal length values to create multiple pseudo-domains and construct an adversarial training loss to encourage the feature representation to be more domain-agnostic. Without bells and whistles, our approach, namely DG-BEV, successfully alleviates the performance drop on the unseen target domain without impairing the accuracy of the source domain. Extensive experiments on various public datasets, including Waymo, nuScenes, and Lyft, demonstrate the generalization and effectiveness of our approach. To the best of our knowledge, this is the first systematic study to explore a domain generalization method for MV3D-Det.

cs.CV↗

arXiv2023

An open unified deep graph learning framework for discovering drug leads

Computational discovery of ideal lead compounds is a critical process for modern drug discovery. It comprises multiple stages: hit screening, molecular property prediction, and molecule optimization. Current efforts are disparate, involving the establishment of models for each stage, followed by multi-stage multi-model integration. However, this is non-ideal, as clumsy integration of incompatible models increases research overheads, and may even reduce success rates in drug discovery. Facilitating compatibilities requires establishing inherent model consistencies across lead discovery stages. Towards that effect, we propose an open deep graph learning (DGL) based pipeline: generative adversarial feature subspace enhancement (GAFSE), which first unifies the modeling of these stages into one learning framework. GAFSE also offers standardized modular design and streamlined interfaces for future expansions and community support. GAFSE combines adversarial/generative learning, graph attention network, graph reconstruction network, and optimizes the classification/regression loss, adversarial/generative loss, and reconstruction loss simultaneously. Convergence analysis theoretically guarantees model generalization performance. Exhaustive benchmarking demonstrates that the GAFSE pipeline achieves excellent performance across almost all lead discovery stages, while also providing valuable model interpretability. Hence, we believe this tool will enhance the efficiency and productivity of drug discovery researchers.

q-bio.BM↗

arXiv2023

Sum-Rate Maximization in Active RIS-Assisted Multi-Antenna WPCN

In this paper, we propose an active reconfigurable intelligent surface (RIS) enabled hybrid relaying scheme for a multi-antenna wireless powered communication network (WPCN), where the active RIS is employed to assist both wireless energy transfer (WET) from the power station (PS) to energy-constrained users and wireless information transmission (WIT) from users to the receiving station (RS). For further performance enhancement, we propose to employ both transmit beamforming at the PS and receive beamforming at the RS. We formulate a sum-rate maximization problem by jointly optimizing the RIS phase shifts and amplitude reflection coefficients for both the WET and the WIT, transmit and receive beamforming vectors, and network resource allocation. To solve this non-convex problem, we propose an efficient alternating optimization algorithm with linear minimum mean squared error criterion, semi-definite relaxation (SDR) and successive convex approximation techniques. Specifically, the tightness of applying the SDR is proved. Simulation results demonstrate that our proposed scheme with 10 reflecting elements (REs) and 4 antennas can achieve 17.78% and 415.48% performance gains compared to the single-antenna scheme with 10 REs and passive RIS scheme with 100 REs, respectively.

cs.IT↗

arXiv2022

Ultra-Low-Frequency Radio Astronomy Observations from a Selenocentric Orbit: first results of the Longjiang-2 experiment

This paper introduces the first results of observations with the Ultra-Long-Wavelength (ULW) -- Low Frequency Interferometer and Spectrometer (LFIS) on board the selenocentric satellite Longjiang-2. We present a brief description of the satellite and focus on the LFIS payload. The in-orbit commissioning confirmed a reliable operational status of the instrumentation. We also present results of a transition observation, which offers unique measurements on several novel aspects. We estimate the RFI suppression required for such a radio astronomy instrumentation at the Moon distances from Earth to be of the order of 80 dB. We analyse a method of separating Earth- and satellite-originated radio frequency interference (RFI). It is found that the RFI level at frequencies lower than a few MHz is smaller than the receiver noise floor.

astro-ph.IM↗

arXiv2022

Findings of the WMT 2022 Shared Task on Translation Suggestion

We report the result of the first edition of the WMT shared task on Translation Suggestion (TS). The task aims to provide alternatives for specific words or phrases given the entire documents generated by machine translation (MT). It consists two sub-tasks, namely, the naive translation suggestion and translation suggestion with hints. The main difference is that some hints are provided in sub-task two, therefore, it is easier for the model to generate more accurate suggestions. For sub-task one, we provide the corpus for the language pairs English-German and English-Chinese. And only English-Chinese corpus is provided for the sub-task two. We received 92 submissions from 5 participating teams in sub-task one and 6 submissions for the sub-task 2, most of them covering all of the translation directions. We used the automatic metric BLEU for evaluating the performance of each submission.

cs.CL↗

arXiv2022

WeTS: A Benchmark for Translation Suggestion

Translation Suggestion (TS), which provides alternatives for specific words or phrases given the entire documents translated by machine translation (MT) \cite{lee2021intellicat}, has been proven to play a significant role in post editing (PE). However, there is still no publicly available data set to support in-depth research for this problem, and no reproducible experimental results can be followed by researchers in this community. To break this limitation, we create a benchmark data set for TS, called \emph{WeTS}, which contains golden corpus annotated by expert translators on four translation directions. Apart from the human-annotated golden corpus, we also propose several novel methods to generate synthetic corpus which can substantially improve the performance of TS. With the corpus we construct, we introduce the Transformer-based model for TS, and experimental results show that our model achieves State-Of-The-Art (SOTA) results on all four translation directions, including English-to-German, German-to-English, Chinese-to-English and English-to-Chinese. Codes and corpus can be found at https://github.com/ZhenYangIACAS/WeTS.git.

cs.CL↗

arXiv2022

Semi-supervised Semantic Segmentation with Prototype-based Consistency Regularization

Semi-supervised semantic segmentation requires the model to effectively propagate the label information from limited annotated images to unlabeled ones. A challenge for such a per-pixel prediction task is the large intra-class variation, i.e., regions belonging to the same class may exhibit a very different appearance even in the same picture. This diversity will make the label propagation hard from pixels to pixels. To address this problem, we propose a novel approach to regularize the distribution of within-class features to ease label propagation difficulty. Specifically, our approach encourages the consistency between the prediction from a linear predictor and the output from a prototype-based predictor, which implicitly encourages features from the same pseudo-class to be close to at least one within-class prototype while staying far from the other between-class prototypes. By further incorporating CutMix operations and a carefully-designed prototype maintenance strategy, we create a semi-supervised semantic segmentation algorithm that demonstrates superior performance over the state-of-the-art methods from extensive experimental evaluation on both Pascal VOC and Cityscapes benchmarks.

cs.CV↗

arXiv2022

Cross-Align: Modeling Deep Cross-lingual Interactions for Word Alignment

Word alignment which aims to extract lexicon translation equivalents between source and target sentences, serves as a fundamental tool for natural language processing. Recent studies in this area have yielded substantial improvements by generating alignments from contextualized embeddings of the pre-trained multilingual language models. However, we find that the existing approaches capture few interactions between the input sentence pairs, which degrades the word alignment quality severely, especially for the ambiguous words in the monolingual context. To remedy this problem, we propose Cross-Align to model deep interactions between the input sentence pairs, in which the source and target sentences are encoded separately with the shared self-attention modules in the shallow layers, while cross-lingual interactions are explicitly constructed by the cross-attention modules in the upper layers. Besides, to train our model effectively, we propose a two-stage training framework, where the model is trained with a simple Translation Language Modeling (TLM) objective in the first stage and then finetuned with a self-supervised alignment objective in the second stage. Experiments show that the proposed Cross-Align achieves the state-of-the-art (SOTA) performance on four out of five language pairs.

cs.CL↗

arXiv2022

Joint Reconfigurable Intelligent Surface Location and Passive Beamforming Optimization for Maximizing the Secrecy-Rate

The physical layer security (PLS) is investigated for reconfigurable intelligent surface (RIS) assisted wireless networks, where a source transmits its confidential information to a legitimate destination with the aid of a single small RIS in the presence of a malicious eavesdropper. A new joint RIS location and passive beamforming (J-LPB) optimization scheme is proposed for the sake of maximizing the secrecy rate under the RIS location constraint and the constraint that the modulus of the reflecting coefficient at each RIS's unit is not larger than 1. Specifically, we analyze the optimal location of the RIS, and conclude that the product involving the source-RIS distance and the RIS-destination distance should be minimized. Since the product minimization problem is nonconvex, we then propose a two-tier optimization algorithm for solving it. Based on the near-optimal RIS 3D location obtained, we further formulate the passive beamforming optimization problem, and then propose to apply the Charnes-Cooper transformation along with the sequential rank-one constraint relaxation (SROCR) algorithm to solve it. Our numerical results show that the secrecy rate of the proposed J-LPB optimization scheme is higher than that of the benchmarks. Explicitly, we use the following benchmarks: the near-source-based RIS location and passive beamforming (NSB-LPB) optimization scheme, the near-destination-based RIS location and passive beamforming (NDB-LPB) optimization scheme, and the random RIS location and passive beamforming (R-LPB) optimization scheme. Finally, the benefits of our J-LPB scheme are further increased with the number of RIS units.

cs.IT↗