Search arXiv⌕ Search

arXiv subjects

Kai Wang

Publications and source records attributed to Kai Wang.

At least 577 records · Page 32Linked to original sources
arXiv2023

Chip-scale, CMOS-compatible, high energy passively Q-switched laser

Chip-scale, high-energy optical pulse generation is becoming increasingly important as we expand activities into hard to reach areas such as space and deep ocean. Q-switching of the laser cavity is the best known technique for generating high-energy pulses, and typically such systems are in the realm of large bench-top solid-state lasers and fiber lasers, especially in the long wavelength range >1.8 um, thanks to their large energy storage capacity. However, in integrated photonics, the very property of tight mode confinement, that enables a small form factor, becomes an impediment to high energy application due to small optical mode cross-section. In this work, we demonstrate complementary metal-oxide-semiconductor (CMOS) compatible, rare-earth gain based large mode area (LMA) passively Q-switched laser in a compact footprint. We demonstrate high on-chip output pulse energy of >150 nJ in single transverse fundamental mode in the eye-safe window (1.9 um), with a slope efficiency ~ 40% in a footprint of ~9 mm2. The high energy pulse generation demonstrated in this work is comparable or in many cases exceeds Q-switched fiber lasers. This bodes well for field applications in medicine and space.

physics.optics↗
arXiv2023

Minimizing the Influence of Misinformation via Vertex Blocking

Information cascade in online social networks can be rather negative, e.g., the spread of rumors may trigger panic. To limit the influence of misinformation in an effective and efficient manner, the influence minimization (IMIN) problem is studied in the literature: given a graph G and a seed set S, blocking at most b vertices such that the influence spread of the seed set is minimized. In this paper, we are the first to prove the IMIN problem is NP-hard and hard to approximate. Due to the hardness of the problem, existing works resort to greedy solutions and use Monte-Carlo Simulations to solve the problem. However, they are cost-prohibitive on large graphs since they have to enumerate all the candidate blockers and compute the decrease of expected spread when blocking each of them. To improve the efficiency, we propose the AdvancedGreedy algorithm (AG) based on a new graph sampling technique that applies the dominator tree structure, which can compute the decrease of the expected spread of all candidate blockers at once. Besides, we further propose the GreedyReplace algorithm (GR) by considering the relationships among candidate blockers. Extensive experiments on 8 real-life graphs demonstrate that our AG and GR algorithms are significantly faster than the state-of-the-art by up to 6 orders of magnitude, and GR can achieve better effectiveness with its time cost close to AG.

cs.DB↗
arXiv2023

Artificial Intelligence System for Detection and Screening of Cardiac Abnormalities using Electrocardiogram Images

The artificial intelligence (AI) system has achieved expert-level performance in electrocardiogram (ECG) signal analysis. However, in underdeveloped countries or regions where the healthcare information system is imperfect, only paper ECGs can be provided. Analysis of real-world ECG images (photos or scans of paper ECGs) remains challenging due to complex environments or interference. In this study, we present an AI system developed to detect and screen cardiac abnormalities (CAs) from real-world ECG images. The system was evaluated on a large dataset of 52,357 patients from multiple regions and populations across the world. On the detection task, the AI system obtained area under the receiver operating curve (AUC) of 0.996 (hold-out test), 0.994 (external test 1), 0.984 (external test 2), and 0.979 (external test 3), respectively. Meanwhile, the detection results of AI system showed a strong correlation with the diagnosis of cardiologists (cardiologist 1 (R=0.794, p<1e-3), cardiologist 2 (R=0.812, p<1e-3)). On the screening task, the AI system achieved AUCs of 0.894 (hold-out test) and 0.850 (external test). The screening performance of the AI system was better than that of the cardiologists (AI system (0.846) vs. cardiologist 1 (0.520) vs. cardiologist 2 (0.480)). Our study demonstrates the feasibility of an accurate, objective, easy-to-use, fast, and low-cost AI system for CA detection and screening. The system has the potential to be used by healthcare professionals, caregivers, and general users to assess CAs based on real-world ECG images.

cs.CV↗
arXiv2023

Deciphering the Language of Nature: A transformer-based language model for deleterious mutations in proteins

Various machine-learning models, including deep neural network models, have already been developed to predict deleteriousness of missense (non-synonymous) mutations. Potential improvements to the current state of the art, however, may still benefit from a fresh look at the biological problem using more sophisticated self-adaptive machine-learning approaches. Recent advances in the natural language processing field show transformer models-a type of deep neural network-to be particularly powerful at modeling sequence information with context dependence. In this study, we introduce MutFormer, a transformer-based model for the prediction of deleterious missense mutations, which uses reference and mutated protein sequences from the human genome as the primary features. MutFormer takes advantage of a combination of self-attention layers and convolutional layers to learn both long-range and short-range dependencies between amino acid mutations in a protein sequence. In this study, we first pre-trained MutFormer on reference protein sequences and mutated protein sequences resulting from common genetic variants observed in human populations. We next examined different fine-tuning methods to successfully apply the model to deleteriousness prediction of missense mutations. Finally, we evaluated MutFormer's performance on multiple testing data sets. We found that MutFormer showed similar or improved performance over a variety of existing tools, including those that used conventional machine-learning approaches. We conclude that MutFormer successfully considers sequence features that are not explored in previous studies and could potentially complement existing computational predictions or empirically generated functional scores to improve our understanding of disease variants.

q-bio.GN↗
arXiv2023

Relating galaxies across different redshift to study galaxy evolution

We propose a general framework leveraging the halo-galaxy connection to link galaxies observed at different redshift in a statistical way, and use the link to infer the redshift evolution of the galaxy population. Our tests based on hydrodynamic simulations show that our method can accurately recover the stellar mass assembly histories up to $z\sim 3$ for present star-forming and quiescent galaxies down to $10^{10}h^{-1}M_{\odot}$. Applying the method to observational data shows that the stellar mass evolution of the main progenitors of galaxies depends strongly on the properties of descendants, such as stellar mass, halo mass, and star formation states. Galaxies hosted by low-mass groups/halos at the present time have since $z\sim 1.8$ grown their stellar mass $\sim 2.5$ times as fast as those hosted by massive clusters. This dependence on host halo mass becomes much weaker for descendant galaxies with similar star formation states. Star-forming galaxies grow about 2-4 times faster than their quiescent counterparts since $z\sim 1.8$. Both TNG and EAGLE simulations over-predict the progenitor stellar mass at $z>1$, particularly for low-mass descendants.

astro-ph.GA↗
arXiv2023

Bioformer: an efficient transformer language model for biomedical text mining

Pretrained language models such as Bidirectional Encoder Representations from Transformers (BERT) have achieved state-of-the-art performance in natural language processing (NLP) tasks. Recently, BERT has been adapted to the biomedical domain. Despite the effectiveness, these models have hundreds of millions of parameters and are computationally expensive when applied to large-scale NLP applications. We hypothesized that the number of parameters of the original BERT can be dramatically reduced with minor impact on performance. In this study, we present Bioformer, a compact BERT model for biomedical text mining. We pretrained two Bioformer models (named Bioformer8L and Bioformer16L) which reduced the model size by 60% compared to BERTBase. Bioformer uses a biomedical vocabulary and was pre-trained from scratch on PubMed abstracts and PubMed Central full-text articles. We thoroughly evaluated the performance of Bioformer as well as existing biomedical BERT models including BioBERT and PubMedBERT on 15 benchmark datasets of four different biomedical NLP tasks: named entity recognition, relation extraction, question answering and document classification. The results show that with 60% fewer parameters, Bioformer16L is only 0.1% less accurate than PubMedBERT while Bioformer8L is 0.9% less accurate than PubMedBERT. Both Bioformer16L and Bioformer8L outperformed BioBERTBase-v1.1. In addition, Bioformer16L and Bioformer8L are 2-3 fold as fast as PubMedBERT/BioBERTBase-v1.1. Bioformer has been successfully deployed to PubTator Central providing gene annotations over 35 million PubMed abstracts and 5 million PubMed Central full-text articles. We make Bioformer publicly available via https://github.com/WGLab/bioformer, including pre-trained models, datasets, and instructions for downstream use.

cs.CL↗
arXiv2023

Artificial non-Abelian lattice gauge fields for photons in the synthetic frequency dimension

Non-Abelian gauge fields give rise to nontrivial topological physics. Here we develop a scheme to create an arbitrary SU(2) lattice gauge field for photons in the synthetic frequency dimension using an array of dynamically modulated ring resonators. The photon polarization is taken as the spin basis to implement the matrix-valued gauge fields. Using a non-Abelian generalization of the Harper-Hofstadter Hamiltonian as a specific example, we show that the measurement of the steady-state photon amplitudes inside the resonators can reveal the band structures of the Hamiltonian, which show signatures of the underlying non-Abelian gauge field. These results provide opportunities to explore novel topological phenomena associated with non-Abelian lattice gauge fields in photonic systems.

physics.optics↗
arXiv2023

Smoothed Online Combinatorial Optimization Using Imperfect Predictions

Smoothed online combinatorial optimization considers a learner who repeatedly chooses a combinatorial decision to minimize an unknown changing cost function with a penalty on switching decisions in consecutive rounds. We study smoothed online combinatorial optimization problems when an imperfect predictive model is available, where the model can forecast the future cost functions with uncertainty. We show that using predictions to plan for a finite time horizon leads to regret dependent on the total predictive uncertainty and an additional switching cost. This observation suggests choosing a suitable planning window to balance between uncertainty and switching cost, which leads to an online algorithm with guarantees on the upper and lower bounds of the cumulative regret. Empirically, our algorithm shows a significant improvement in cumulative regret compared to other baselines in synthetic online distributed streaming problems.

cs.LG↗
arXiv2023

OneRing: A Simple Method for Source-free Open-partial Domain Adaptation

In this paper, we investigate Source-free Open-partial Domain Adaptation (SF-OPDA), which addresses the situation where there exist both domain and category shifts between source and target domains. Under the SF-OPDA setting, which aims to address data privacy concerns, the model cannot access source data anymore during target adaptation. We propose a novel training scheme to learn a (n+1)-way classifier to predict the n source classes and the unknown class, where samples of only known source categories are available for training. Furthermore, for target adaptation, we simply adopt a weighted entropy minimization to adapt the source pretrained model to the unlabeled target domain without source data. In experiments, we show our simple method surpasses current OPDA approaches which demand source data during adaptation. When augmented with a closed-set domain adaptation approach during target adaptation, our source-free method further outperforms the current state-of-the-art OPDA method by 2.5%, 7.2% and 13% on Office-31, Office-Home and VisDA respectively.

cs.CV↗
arXiv2022

Jet-cloud/star interaction as an interpretation of neutrino outburst from the blazar TXS 0506+056

A neutrino outburst between September 2014 and March 2015 was discovered from the blazar TXS 0506+056 by an investigation of $9.5$ years of IceCube data, while the blazar was in a quiescent state during the outburst with a gamma-ray flux only about one-fifth of the neutrino flux. In this work, we give a possible interpretation of the abnormal feature by proposing that the neutrino outburst originates from the interaction between a relativistic jet and a dense gas cloud formed via the tidally disrupted envelope of a red giant being blown apart by the impact of the jet. Gamma-ray photons and electron/positron pairs that are produced through the hadronuclear interactions correspondingly will induce electromagnetic cascades and then make the cloud ionized and thermalized. The EM radiation from jet-cloud/star interaction is mainly contributed by the relatively low-energy relativistic protons which propagate in the diffusion regime inside the cloud due to magnetic deflections, whereas the observed high-energy neutrinos ($\gtrsim 100\,\rm TeV$) are produced by the relatively high-energy protons which can keep beamed owing to the weak magnetic deflections, inducing a much higher flux of neutrinos than electromagnetic radiation. The observed low-energy electromagnetic radiations during the neutrino outburst period are almost as same as that in the quiescent state of the source, so it may mainly arise as same as the general quiescent state. As a result, due to the intrusion of a dense cloud, the neutrino outburst can be expected, and in the meantime, the accompanying electromagnetic radiations from hadronic processes will not cause any enhancement in the blazar's electromagnetic flux.

astro-ph.HE↗
arXiv2022

FECAM: Frequency Enhanced Channel Attention Mechanism for Time Series Forecasting

Time series forecasting is a long-standing challenge due to the real-world information is in various scenario (e.g., energy, weather, traffic, economics, earthquake warning). However some mainstream forecasting model forecasting result is derailed dramatically from ground truth. We believe it's the reason that model's lacking ability of capturing frequency information which richly contains in real world datasets. At present, the mainstream frequency information extraction methods are Fourier transform(FT) based. However, use of FT is problematic due to Gibbs phenomenon. If the values on both sides of sequences differ significantly, oscillatory approximations are observed around both sides and high frequency noise will be introduced. Therefore We propose a novel frequency enhanced channel attention that adaptively modelling frequency interdependencies between channels based on Discrete Cosine Transform which would intrinsically avoid high frequency noise caused by problematic periodity during Fourier Transform, which is defined as Gibbs Phenomenon. We show that this network generalize extremely effectively across six real-world datasets and achieve state-of-the-art performance, we further demonstrate that frequency enhanced channel attention mechanism module can be flexibly applied to different networks. This module can improve the prediction ability of existing mainstream networks, which reduces 35.99% MSE on LSTM, 10.01% on Reformer, 8.71% on Informer, 8.29% on Autoformer, 8.06% on Transformer, etc., at a slight computational cost ,with just a few line of code. Our codes and data are available at https://github.com/Zero-coder/FECAM.

cs.AI↗
arXiv2022

Self adaptive global-local feature enhancement for radiology report generation

Automated radiology report generation aims at automatically generating a detailed description of medical images, which can greatly alleviate the workload of radiologists and provide better medical services to remote areas. Most existing works pay attention to the holistic impression of medical images, failing to utilize important anatomy information. However, in actual clinical practice, radiologists usually locate important anatomical structures, and then look for signs of abnormalities in certain structures and reason the underlying disease. In this paper, we propose a novel framework AGFNet to dynamically fuse the global and anatomy region feature to generate multi-grained radiology report. Firstly, we extract important anatomy region features and global features of input Chest X-ray (CXR). Then, with the region features and the global features as input, our proposed self-adaptive fusion gate module could dynamically fuse multi-granularity information. Finally, the captioning generator generates the radiology reports through multi-granularity features. Experiment results illustrate that our model achieved the state-of-the-art performance on two benchmark datasets including the IU X-Ray and MIMIC-CXR. Further analyses also prove that our model is able to leverage the multi-grained information from radiology images and texts so as to help generate more accurate reports.

cs.CV↗
arXiv2022

Decision-Focused Learning without Differentiable Optimization: Learning Locally Optimized Decision Losses

Decision-Focused Learning (DFL) is a paradigm for tailoring a predictive model to a downstream optimization task that uses its predictions in order to perform better on that specific task. The main technical challenge associated with DFL is that it requires being able to differentiate through the optimization problem, which is difficult due to discontinuous solutions and other challenges. Past work has largely gotten around this issue by handcrafting task-specific surrogates to the original optimization problem that provide informative gradients when differentiated through. However, the need to handcraft surrogates for each new task limits the usability of DFL. In addition, there are often no guarantees about the convexity of the resulting surrogates and, as a result, training a predictive model using them can lead to inferior local optima. In this paper, we do away with surrogates altogether and instead learn loss functions that capture task-specific information. To the best of our knowledge, ours is the first approach that entirely replaces the optimization component of decision-focused learning with a loss that is automatically learned. Our approach (a) only requires access to a black-box oracle that can solve the optimization problem and is thus generalizable, and (b) can be convex by construction and so can be easily optimized over. We evaluate our approach on three resource allocation problems from the literature and find that our approach outperforms learning without taking into account task structure in all three domains, and even hand-crafted surrogates from the literature.

cs.LG↗
arXiv2022

Dataset Distillation via Factorization

In this paper, we study \xw{dataset distillation (DD)}, from a novel perspective and introduce a \emph{dataset factorization} approach, termed \emph{HaBa}, which is a plug-and-play strategy portable to any existing DD baseline. Unlike conventional DD approaches that aim to produce distilled and representative samples, \emph{HaBa} explores decomposing a dataset into two components: data \emph{Ha}llucination networks and \emph{Ba}ses, where the latter is fed into the former to reconstruct image samples. The flexible combinations between bases and hallucination networks, therefore, equip the distilled data with exponential informativeness gain, which largely increase the representation capability of distilled datasets. To furthermore increase the data efficiency of compression results, we further introduce a pair of adversarial contrastive constraints on the resultant hallucination networks and bases, which increase the diversity of generated images and inject more discriminant information into the factorization. Extensive comparisons and experiments demonstrate that our method can yield significant improvement on downstream classification tasks compared with previous state of the arts, while reducing the total number of compressed parameters by up to 65\%. Moreover, distilled datasets by our approach also achieve \textasciitilde10\% higher accuracy than baseline methods in cross-architecture generalization. Our code is available \href{https://github.com/Huage001/DatasetFactorization}{here}.

cs.CV↗
arXiv2022

Point-to-Box Network for Accurate Object Detection via Single Point Supervision

Object detection using single point supervision has received increasing attention over the years. However, the performance gap between point supervised object detection (PSOD) and bounding box supervised detection remains large. In this paper, we attribute such a large performance gap to the failure of generating high-quality proposal bags which are crucial for multiple instance learning (MIL). To address this problem, we introduce a lightweight alternative to the off-the-shelf proposal (OTSP) method and thereby create the Point-to-Box Network (P2BNet), which can construct an inter-objects balanced proposal bag by generating proposals in an anchor-like way. By fully investigating the accurate position information, P2BNet further constructs an instance-level bag, avoiding the mixture of multiple objects. Finally, a coarse-to-fine policy in a cascade fashion is utilized to improve the IoU between proposals and ground-truth (GT). Benefiting from these strategies, P2BNet is able to produce high-quality instance-level bags for object detection. P2BNet improves the mean average precision (AP) by more than 50% relative to the previous best PSOD method on the MS COCO dataset. It also demonstrates the great potential to bridge the performance gap between point supervised and bounding-box supervised detectors. The code will be released at github.com/ucas-vg/P2BNet.

cs.CV↗
arXiv2022

MV-HAN: A Hybrid Attentive Networks based Multi-View Learning Model for Large-scale Contents Recommendation

Industrial recommender systems usually employ multi-source data to improve the recommendation quality, while effectively sharing information between different data sources remain a challenge. In this paper, we introduce a novel Multi-View Approach with Hybrid Attentive Networks (MV-HAN) for contents retrieval at the matching stage of recommender systems. The proposed model enables high-order feature interaction from various input features while effectively transferring knowledge between different types. By employing a well-placed parameters sharing strategy, the MV-HAN substantially improves the retrieval performance in sparse types. The designed MV-HAN inherits the efficiency advantages in the online service from the two-tower model, by mapping users and contents of different types into the same features space. This enables fast retrieval of similar contents with an approximate nearest neighbor algorithm. We conduct offline experiments on several industrial datasets, demonstrating that the proposed MV-HAN significantly outperforms baselines on the content retrieval tasks. Importantly, the MV-HAN is deployed in a real-world matching system. Online A/B test results show that the proposed method can significantly improve the quality of recommendations.

cs.IR↗
arXiv2022

Positive Pair Distillation Considered Harmful: Continual Meta Metric Learning for Lifelong Object Re-Identification

Lifelong object re-identification incrementally learns from a stream of re-identification tasks. The objective is to learn a representation that can be applied to all tasks and that generalizes to previously unseen re-identification tasks. The main challenge is that at inference time the representation must generalize to previously unseen identities. To address this problem, we apply continual meta metric learning to lifelong object re-identification. To prevent forgetting of previous tasks, we use knowledge distillation and explore the roles of positive and negative pairs. Based on our observation that the distillation and metric losses are antagonistic, we propose to remove positive pairs from distillation to robustify model updates. Our method, called Distillation without Positive Pairs (DwoPP), is evaluated on extensive intra-domain experiments on person and vehicle re-identification datasets, as well as inter-domain experiments on the LReID benchmark. Our experiments demonstrate that DwoPP significantly outperforms the state-of-the-art. The code is here: https://github.com/wangkai930418/DwoPP_code

cs.CV↗