Searcharxiv⌕ Search

arXiv subjects

Hao Chen

Publications and source records attributed to Hao Chen.

At least 955 records · Page 53Linked to original sources
arXiv2023

FRB 20210405I: a nearby Fast Radio Burst localised to sub-arcsecond precision with MeerKAT

We present the first sub-arcsecond localised Fast Radio Burst (FRB) detected using MeerKAT. FRB 20210405I was detected in the incoherent beam using the MeerTRAP pipeline on 2021 April 05 with a signal to noise ratio of 140.8 and a dispersion measure of 565.17 pc cm$^{-3}$. It was detected while MeerTRAP was observing commensally with the ThunderKAT large survey project, and was sufficiently bright that we could use the ThunderKAT 8s images to localise the FRB. Two different models of the dispersion measure in the Milky Way and halo suggest that the source is either right at the edge of the Galaxy, or outside. This highlights the uncertainty in the Milky Way dispersion measure models, particularly in the Galactic Plane, and the uncertainty of Milky Way halo models. Further investigation and modelling of these uncertainties will be facilitated by future detections and localisations of nearby FRBs. We use the combined localisation, dispersion measure, scattering, specific luminosity and chance coincidence probability information to find that the origin is most likely extra-galactic and identify the likely host galaxy of the FRB: 2MASS J1701249$-$4932475. Using SALT spectroscopy and archival observations of the field, we find that the host is a disk/spiral galaxy at a redshift of $z=0.066$.

astro-ph.HE↗
arXiv2023

Train Once, Get a Family: State-Adaptive Balances for Offline-to-Online Reinforcement Learning

Offline-to-online reinforcement learning (RL) is a training paradigm that combines pre-training on a pre-collected dataset with fine-tuning in an online environment. However, the incorporation of online fine-tuning can intensify the well-known distributional shift problem. Existing solutions tackle this problem by imposing a policy constraint on the policy improvement objective in both offline and online learning. They typically advocate a single balance between policy improvement and constraints across diverse data collections. This one-size-fits-all manner may not optimally leverage each collected sample due to the significant variation in data quality across different states. To this end, we introduce Family Offline-to-Online RL (FamO2O), a simple yet effective framework that empowers existing algorithms to determine state-adaptive improvement-constraint balances. FamO2O utilizes a universal model to train a family of policies with different improvement/constraint intensities, and a balance model to select a suitable policy for each state. Theoretically, we prove that state-adaptive balances are necessary for achieving a higher policy performance upper bound. Empirically, extensive experiments show that FamO2O offers a statistically significant improvement over various existing methods, achieving state-of-the-art performance on the D4RL benchmark. Codes are available at https://github.com/LeapLabTHU/FamO2O.

cs.LG↗
arXiv2023

Searching for axion dark matter with MeerKAT Radio Telescope

Axions provide a natural and well-motivated dark matter candidate, with the capability to convert directly to photons in the presence of an electromagnetic field. A particularly compelling observational target is the conversion of dark matter axions into photons in the magnetospheres of highly magnetised neutron stars, which is expected to produce a narrow spectral peak centred at the frequency of the axion mass. We point the MeerKAT radio telescope towards the isolated neutron star J0806.4$-$4123 for $10$-hours of observation and obtain the radio spectra in the frequency range $769$-$1051$ MHz. By modelling the conversion process of infalling axion dark matter (DM), we then compare these spectra to theoretical expectations for a given choice of axion parameters. Whilst finding no signal above $5σ$ in the data, we provide a unique constraint on the Primakoff coupling of axion DM, $g_{{\rm a}γγ}\lesssim 9.3 \times 10^{-12}\,{\rm GeV}^{-1}$ at the $95\%$ confidence level, in the mass range $3.18$-$4.35\,μ$eV. This result serves the strongest constraint in the axion mass range $4.20$-$4.35\,μ$eV.

astro-ph.CO↗
arXiv2023

Customising General Large Language Models for Specialised Emotion Recognition Tasks

The advent of large language models (LLMs) has gained tremendous attention over the past year. Previous studies have shown the astonishing performance of LLMs not only in other tasks but also in emotion recognition in terms of accuracy, universality, explanation, robustness, few/zero-shot learning, and others. Leveraging the capability of LLMs inevitably becomes an essential solution for emotion recognition. To this end, we further comprehensively investigate how LLMs perform in linguistic emotion recognition if we concentrate on this specific task. Specifically, we exemplify a publicly available and widely used LLM -- Chat General Language Model, and customise it for our target by using two different modal adaptation techniques, i.e., deep prompt tuning and low-rank adaptation. The experimental results obtained on six widely used datasets present that the adapted LLM can easily outperform other state-of-the-art but specialised deep models. This indicates the strong transferability and feasibility of LLMs in the field of emotion recognition.

cs.CL↗
arXiv2023

Continuous Cross-resolution Remote Sensing Image Change Detection

Most contemporary supervised Remote Sensing (RS) image Change Detection (CD) approaches are customized for equal-resolution bitemporal images. Real-world applications raise the need for cross-resolution change detection, aka, CD based on bitemporal images with different spatial resolutions. Given training samples of a fixed bitemporal resolution difference (ratio) between the high-resolution (HR) image and the low-resolution (LR) one, current cross-resolution methods may fit a certain ratio but lack adaptation to other resolution differences. Toward continuous cross-resolution CD, we propose scale-invariant learning to enforce the model consistently predicting HR results given synthesized samples of varying resolution differences. Concretely, we synthesize blurred versions of the HR image by random downsampled reconstructions to reduce the gap between HR and LR images. We introduce coordinate-based representations to decode per-pixel predictions by feeding the coordinate query and corresponding multi-level embedding features into an MLP that implicitly learns the shape of land cover changes, therefore benefiting recognizing blurred objects in the LR image. Moreover, considering that spatial resolution mainly affects the local textures, we apply local-window self-attention to align bitemporal features during the early stages of the encoder. Extensive experiments on two synthesized and one real-world different-resolution CD datasets verify the effectiveness of the proposed method. Our method significantly outperforms several vanilla CD methods and two cross-resolution CD methods on the three datasets both in in-distribution and out-of-distribution settings. The empirical results suggest that our method could yield relatively consistent HR change predictions regardless of varying bitemporal resolution ratios. Our code is available at \url{https://github.com/justchenhao/SILI_CD}.

cs.CV↗
arXiv2023

De novo protein design using geometric vector field networks

Innovations like protein diffusion have enabled significant progress in de novo protein design, which is a vital topic in life science. These methods typically depend on protein structure encoders to model residue backbone frames, where atoms do not exist. Most prior encoders rely on atom-wise features, such as angles and distances between atoms, which are not available in this context. Thus far, only several simple encoders, such as IPA, have been proposed for this scenario, exposing the frame modeling as a bottleneck. In this work, we proffer the Vector Field Network (VFN), which enables network layers to perform learnable vector computations between coordinates of frame-anchored virtual atoms, thus achieving a higher capability for modeling frames. The vector computation operates in a manner similar to a linear layer, with each input channel receiving 3D virtual atom coordinates instead of scalar values. The multiple feature vectors output by the vector computation are then used to update the residue representations and virtual atom coordinates via attention aggregation. Remarkably, VFN also excels in modeling both frames and atoms, as the real atoms can be treated as the virtual atoms for modeling, positioning VFN as a potential universal encoder. In protein diffusion (frame modeling), VFN exhibits an impressive performance advantage over IPA, excelling in terms of both designability (67.04% vs. 53.58%) and diversity (66.54% vs. 51.98%). In inverse folding (frame and atom modeling), VFN outperforms the previous SoTA model, PiFold (54.7% vs. 51.66%), on sequence recovery rate. We also propose a method of equipping VFN with the ESM model, which significantly surpasses the previous ESM-based SoTA (62.67% vs. 55.65%), LM-Design, by a substantial margin.

cs.CE↗
arXiv2023

Towards Intelligent Network Management: Leveraging AI for Network Service Detection

As the complexity and scale of modern computer networks continue to increase, there has emerged an urgent need for precise traffic analysis, which plays a pivotal role in cutting-edge wireless connectivity technologies. This study focuses on leveraging Machine Learning methodologies to create an advanced network traffic classification system. We introduce a novel data-driven approach that excels in identifying various network service types in real-time, by analyzing patterns within the network traffic. Our method organizes similar kinds of network traffic into distinct categories, referred to as network services, based on latency requirement. Furthermore, it decomposes the network traffic stream into multiple, smaller traffic flows, with each flow uniquely carrying a specific service. Our ML models are trained on a dataset comprised of labeled examples representing different network service types collected on various Wi-Fi network conditions. Upon evaluation, our system demonstrates a remarkable accuracy in distinguishing the network services. These results emphasize the substantial promise of integrating Artificial Intelligence in wireless technologies. Such an approach encourages more efficient energy consumption, enhances Quality of Service assurance, and optimizes the allocation of network resources, thus laying a solid groundwork for the development of advanced intelligent networks.

cs.NI↗
arXiv2023

Catenoid limits of singly periodic minimal surfaces with Scherk-type ends

We construct families of embedded, singly periodic minimal surfaces of any genus $g$ in the quotient with any even number $2n>2$ of almost parallel Scherk ends. A surface in such a family looks like $n$ parallel planes connected by $n-1+g$ small catenoid necks. In the limit, the family converges to an $n$-sheeted vertical plane with $n-1+g$ singular points termed nodes in the quotient. For the nodes to open up into catenoid necks, their locations must satisfy a set of balance equations whose solutions are given by the roots of Stieltjes polynomials.

math.DG↗
arXiv2023

Gluing Karcher-Scherk saddle towers II: Singly periodic minimal surfaces

This is the second in a series of papers that construct minimal surfaces by gluing singly periodic Karcher--Scherk saddle towers along their wings. This paper aims to construct singly periodic minimal surfaces with Scherk ends. As in the first paper, we prescribe phase differences between saddle towers, and obtain many new examples without any horizontal reflection plane. This construction is not very different from previous ones, hence we will only provide sketched proofs. We will however study the embeddedness with great care. Previously, embeddedness can not be determined in the presence of "parallel" Scherk ends, as it was not clear if they bend towards or away from each other. In a recent study, the bending was completely ignored and embeddedness was falsely claimed. We correct this mistake by carefully analysing slight bendings, thus identify scenarios where the constructed surfaces are indeed embedded.

math.DG↗
arXiv2023

DatasetDM: Synthesizing Data with Perception Annotations Using Diffusion Models

Current deep networks are very data-hungry and benefit from training on largescale datasets, which are often time-consuming to collect and annotate. By contrast, synthetic data can be generated infinitely using generative models such as DALL-E and diffusion models, with minimal effort and cost. In this paper, we present DatasetDM, a generic dataset generation model that can produce diverse synthetic images and the corresponding high-quality perception annotations (e.g., segmentation masks, and depth). Our method builds upon the pre-trained diffusion model and extends text-guided image synthesis to perception data generation. We show that the rich latent code of the diffusion model can be effectively decoded as accurate perception annotations using a decoder module. Training the decoder only needs less than 1% (around 100 images) manually labeled images, enabling the generation of an infinitely large annotated dataset. Then these synthetic data can be used for training various perception models for downstream tasks. To showcase the power of the proposed approach, we generate datasets with rich dense pixel-wise labels for a wide range of downstream tasks, including semantic segmentation, instance segmentation, and depth estimation. Notably, it achieves 1) state-of-the-art results on semantic segmentation and instance segmentation; 2) significantly more robust on domain generalization than using the real data alone; and state-of-the-art results in zero-shot segmentation setting; and 3) flexibility for efficient application and novel task composition (e.g., image editing). The project website and code can be found at https://weijiawu.github.io/DatasetDM_page/ and https://github.com/showlab/DatasetDM, respectively

cs.CV↗
arXiv2023

ChatRadio-Valuer: A Chat Large Language Model for Generalizable Radiology Report Generation Based on Multi-institution and Multi-system Data

Radiology report generation, as a key step in medical image analysis, is critical to the quantitative analysis of clinically informed decision-making levels. However, complex and diverse radiology reports with cross-source heterogeneity pose a huge generalizability challenge to the current methods under massive data volume, mainly because the style and normativity of radiology reports are obviously distinctive among institutions, body regions inspected and radiologists. Recently, the advent of large language models (LLM) offers great potential for recognizing signs of health conditions. To resolve the above problem, we collaborate with the Second Xiangya Hospital in China and propose ChatRadio-Valuer based on the LLM, a tailored model for automatic radiology report generation that learns generalizable representations and provides a basis pattern for model adaptation in sophisticated analysts' cases. Specifically, ChatRadio-Valuer is trained based on the radiology reports from a single institution by means of supervised fine-tuning, and then adapted to disease diagnosis tasks for human multi-system evaluation (i.e., chest, abdomen, muscle-skeleton, head, and maxillofacial $\&$ neck) from six different institutions in clinical-level events. The clinical dataset utilized in this study encompasses a remarkable total of \textbf{332,673} observations. From the comprehensive results on engineering indicators, clinical efficacy and deployment cost metrics, it can be shown that ChatRadio-Valuer consistently outperforms state-of-the-art models, especially ChatGPT (GPT-3.5-Turbo) and GPT-4 et al., in terms of the diseases diagnosis from radiology reports. ChatRadio-Valuer provides an effective avenue to boost model generalization performance and alleviate the annotation workload of experts to enable the promotion of clinical AI applications in radiology reports.

cs.CL↗
arXiv2023

The Program Testing Ability of Large Language Models for Code

Recent development of large language models (LLMs) for code like CodeX and CodeT5+ demonstrates tremendous promise in achieving code intelligence. Their ability of synthesizing code that completes a program for performing a pre-defined task has been intensively tested and verified on benchmark datasets including HumanEval and MBPP. Yet, evaluation of these LLMs from more perspectives (than just program synthesis) is also anticipated, considering their broad scope of applications in software engineering. In this paper, we explore the ability of LLMs for testing programs/code. By performing thorough analyses of recent LLMs for code in program testing, we show a series of intriguing properties of these models and demonstrate how program testing ability of LLMs can be improved. Following recent work which utilizes generated test cases to enhance program synthesis, we further leverage our findings in improving the quality of the synthesized programs and show +11.77% and +4.22% higher code pass rates on HumanEval+ comparing with the GPT-3.5-turbo baseline and the recent state-of-the-art, respectively.

cs.CL↗
arXiv2023

UBSea: A Unified Community Detection Framework

Detecting communities in networks and graphs is an important task across many disciplines such as statistics, social science and engineering. There are generally three different kinds of mixing patterns for the case of two communities: assortative mixing, disassortative mixing and core-periphery structure. Modularity optimization is a classical way for fitting network models with communities. However, it can only deal with assortative mixing and disassortative mixing when the mixing pattern is known and fails to discover the core-periphery structure. In this paper, we extend modularity in a strategic way and propose a new framework based on Unified Bigroups Standadized Edge-count Analysis (UBSea). It can address all the formerly mentioned community mixing structures. In addition, this new framework is able to automatically choose the mixing type to fit the networks. Simulation studies show that the new framework has superb performance in a wide range of settings under the stochastic block model and the degree-corrected stochastic block model. We show that the new approach produces consistent estimate of the communities under a suitable signal-to-noise-ratio condition, for the case of a block model with two communities, for both undirected and directed networks. The new method is illustrated through applications to several real-world datasets.

stat.ME↗
arXiv2023

Hoeffding's Inequality for Markov Chains under Generalized Concentrability Condition

This paper studies Hoeffding's inequality for Markov chains under the generalized concentrability condition defined via integral probability metric (IPM). The generalized concentrability condition establishes a framework that interpolates and extends the existing hypotheses of Markov chain Hoeffding-type inequalities. The flexibility of our framework allows Hoeffding's inequality to be applied beyond the ergodic Markov chains in the traditional sense. We demonstrate the utility by applying our framework to several non-asymptotic analyses arising from the field of machine learning, including (i) a generalization bound for empirical risk minimization with Markovian samples, (ii) a finite sample guarantee for Ployak-Ruppert averaging of SGD, and (iii) a new regret bound for rested Markovian bandits with general state space.

stat.ML↗
arXiv2023

Towards Domain-Specific Features Disentanglement for Domain Generalization

Distributional shift between domains poses great challenges to modern machine learning algorithms. The domain generalization (DG) signifies a popular line targeting this issue, where these methods intend to uncover universal patterns across disparate distributions. Noted, the crucial challenge behind DG is the existence of irrelevant domain features, and most prior works overlook this information. Motivated by this, we propose a novel contrastive-based disentanglement method CDDG, to effectively utilize the disentangled features to exploit the over-looked domain-specific features, and thus facilitating the extraction of the desired cross-domain category features for DG tasks. Specifically, CDDG learns to decouple inherent mutually exclusive features by leveraging them in the latent space, thus making the learning discriminative. Extensive experiments conducted on various benchmark datasets demonstrate the superiority of our method compared to other state-of-the-art approaches. Furthermore, visualization evaluations confirm the potential of our method in achieving effective feature disentanglement.

cs.CV↗
arXiv2023

Joint Computing, Pushing, and Caching Optimization for Mobile Edge Computing Networks via Soft Actor-Critic Learning

Mobile edge computing (MEC) networks bring computing and storage capabilities closer to edge devices, which reduces latency and improves network performance. However, to further reduce transmission and computation costs while satisfying user-perceived quality of experience, a joint optimization in computing, pushing, and caching is needed. In this paper, we formulate the joint-design problem in MEC networks as an infinite-horizon discounted-cost Markov decision process and solve it using a deep reinforcement learning (DRL)-based framework that enables the dynamic orchestration of computing, pushing, and caching. Through the deep networks embedded in the DRL structure, our framework can implicitly predict user future requests and push or cache the appropriate content to effectively enhance system performance. One issue we encountered when considering three functions collectively is the curse of dimensionality for the action space. To address it, we relaxed the discrete action space into a continuous space and then adopted soft actor-critic learning to solve the optimization problem, followed by utilizing a vector quantization method to obtain the desired discrete action. Additionally, an action correction method was proposed to compress the action space further and accelerate the convergence. Our simulations under the setting of a general single-user, single-server MEC network with dynamic transmission link quality demonstrate that the proposed framework effectively decreases transmission bandwidth and computing cost by proactively pushing data on future demand to users and jointly optimizing the three functions. We also conduct extensive parameter tuning analysis, which shows that our approach outperforms the baselines under various parameter settings.

cs.IT↗
arXiv2023

Human-like Energy Management Based on Deep Reinforcement Learning and Historical Driving Experiences

Development of hybrid electric vehicles depends on an advanced and efficient energy management strategy (EMS). With online and real-time requirements in mind, this article presents a human-like energy management framework for hybrid electric vehicles according to deep reinforcement learning methods and collected historical driving data. The hybrid powertrain studied has a series-parallel topology, and its control-oriented modeling is founded first. Then, the distinctive deep reinforcement learning (DRL) algorithm, named deep deterministic policy gradient (DDPG), is introduced. To enhance the derived power split controls in the DRL framework, the global optimal control trajectories obtained from dynamic programming (DP) are regarded as expert knowledge to train the DDPG model. This operation guarantees the optimality of the proposed control architecture. Moreover, the collected historical driving data based on experienced drivers are employed to replace the DP-based controls, and thus construct the human-like EMSs. Finally, different categories of experiments are executed to estimate the optimality and adaptability of the proposed human-like EMS. Improvements in fuel economy and convergence rate indicate the effectiveness of the constructed control structure.

eess.SY↗
arXiv2023

Decision-making for Autonomous Vehicles on Highway: Deep Reinforcement Learning with Continuous Action Horizon

Decision-making strategy for autonomous vehicles de-scribes a sequence of driving maneuvers to achieve a certain navigational mission. This paper utilizes the deep reinforcement learning (DRL) method to address the continuous-horizon decision-making problem on the highway. First, the vehicle kinematics and driving scenario on the freeway are introduced. The running objective of the ego automated vehicle is to execute an efficient and smooth policy without collision. Then, the particular algorithm named proximal policy optimization (PPO)-enhanced DRL is illustrated. To overcome the challenges in tardy training efficiency and sample inefficiency, this applied algorithm could realize high learning efficiency and excellent control performance. Finally, the PPO-DRL-based decision-making strategy is estimated from multiple perspectives, including the optimality, learning efficiency, and adaptability. Its potential for online application is discussed by applying it to similar driving scenarios.

cs.AI↗