arXiv subjects

Yan Wu

Publications and source records attributed to Yan Wu.

At least 73 records · Page 4Linked to original sources

arXiv2024

Liquid sloshing behaviours in an elastic tank and suppression effect of baffles

In this paper, a fluid-structure interaction (FSI) framework based on the smoothed particle hydrodynamics (SPH) method is employed to investigate the forces and deformations experienced by LNG tanks during liquid sloshing. As a Lagrangian approach, the SPH method offers the advantage of accurately modelling free-surface flow. The fluid phase consisting of water and air is modelled as a multi-phase system for getting closer to real transport situations. Additionally, the application of FSI within a single framework reduces data transfer discrepancies between fluid dynamics and solid mechanics. To validate the reliability of the numerical methodology, the simulation results about the free surface elevation and wave profiles are compared with experimental data. Subsequently, ring baffles and vertical baffles are introduced separately. While the degree of force acting on the tanks is assessed, the anti-sloshing effectiveness of baffles on sloshing suppression and the variations in stress and strain distributions are evaluated. Further, to compare the influence of the material properties of baffles on sloshing phenomena, the rigid baffle and elastic baffle with different Young's moduli are immersed in the liquid. The results indicate that in this LNG tank configuration, the closer the baffle properties align with rigidity, the more effective the sloshing inhibition.

physics.flu-dyn↗

arXiv2024

Robust Training Objectives Improve Embedding-based Retrieval in Industrial Recommendation Systems

Improving recommendation systems (RS) can greatly enhance the user experience across many domains, such as social media. Many RS utilize embedding-based retrieval (EBR) approaches to retrieve candidates for recommendation. In an EBR system, the embedding quality is key. According to recent literature, self-supervised multitask learning (SSMTL) has showed strong performance on academic benchmarks in embedding learning and resulted in an overall improvement in multiple downstream tasks, demonstrating a larger resilience to the adverse conditions between each downstream task and thereby increased robustness and task generalization ability through the training objective. However, whether or not the success of SSMTL in academia as a robust training objectives translates to large-scale (i.e., over hundreds of million users and interactions in-between) industrial RS still requires verification. Simply adopting academic setups in industrial RS might entail two issues. Firstly, many self-supervised objectives require data augmentations (e.g., embedding masking/corruption) over a large portion of users and items, which is prohibitively expensive in industrial RS. Furthermore, some self-supervised objectives might not align with the recommendation task, which might lead to redundant computational overheads or negative transfer. In light of these two challenges, we evaluate using a robust training objective, specifically SSMTL, through a large-scale friend recommendation system on a social media platform in the tech sector, identifying whether this increase in robustness can work at scale in enhancing retrieval in the production setting. Through online A/B testing with SSMTL-based EBR, we observe statistically significant increases in key metrics in the friend recommendations, with up to 5.45% improvements in new friends made and 1.91% improvements in new friends made with cold-start users.

cs.IR↗

arXiv2024

Learning Stable Robot Grasping with Transformer-based Tactile Control Policies

Measuring grasp stability is an important skill for dexterous robot manipulation tasks, which can be inferred from haptic information with a tactile sensor. Control policies have to detect rotational displacement and slippage from tactile feedback, and determine a re-grasp strategy in term of location and force. Classic stable grasp task only trains control policies to solve for re-grasp location with objects of fixed center of gravity. In this work, we propose a revamped version of stable grasp task that optimises both re-grasp location and gripping force for objects with unknown and moving center of gravity. We tackle this task with a model-free, end-to-end Transformer-based reinforcement learning framework. We show that our approach is able to solve both objectives after training in both simulation and in a real-world setup with zero-shot transfer. We also provide performance analysis of different models to understand the dynamics of optimizing two opposing objectives.

cs.RO↗

arXiv2024

Spin waves in Dirac semimetal Ca$_{0.6}$Sr$_{0.4}$MnSb$_2$ investigated with neutrons by the diffraction method

We report neutron diffraction measurements of Ca$_{0.6}$Sr$_{0.4}$MnSb$_2$, a low-carrier-density Dirac semimetal in which the antiferromagnetic Mn layers are interleaved with Sb layers that host Dirac fermions. We have discovered that we can detect a good quality inelastic spin wave signal from a small (m ~ 0.28 g) single crystal sample by the diffraction method, without energy analysis, using a neutron diffractometer with a position-sensitive area detector; the spin-waves appear as diffuse scattering that is shaped by energy-momentum conservation. By fitting this characteristic magnetic scattering to a spin-wave model, we refine all parameters of the model spin Hamiltonian, including the inter-plane interaction, through use of a three-dimensional measurement in reciprocal space. We also measure the temperature dependence of the spin waves, including the softening of the spin gap on approaching the Neel temperature, $T_N$. Not only do our results provide important new insights into an interplay of magnetism and Dirac electrons, they also establish a new, high-throughput approach to characterizing magnetic excitations on a modern diffractometer without direct energy analysis. Our work opens exciting new opportunities for the follow-up parametric and compositional studies on small, ~0.1 g crystals.

cond-mat.str-el↗

arXiv2024

Data-Driven Symbol Detection for Intersymbol Interference Channels with Bursty Impulsive Noise

We developed machine learning approaches for data-driven trellis-based soft symbol detection in coded transmission over intersymbol interference (ISI) channels in presence of bursty impulsive noise (IN), for example encountered in wireless digital broadcasting systems and vehicular communications. This enabled us to obtain optimized detectors based on the Bahl-Cocke-Jelinek-Raviv (BCJR) algorithm while circumventing the use of full channel state information (CSI) for computing likelihoods and trellis state transition probabilities. First, we extended the application of the neural network (NN)-aided BCJR, recently proposed for ISI channels with additive white Gaussian noise (AWGN). Although suitable for estimating likelihoods via labeling of transmission sequences, the BCJR-NN method does not provide a framework for learning the trellis state transitions. In addition to detection over the joint ISI and IN states we also focused on another scenario where trellis transitions are not trivial: detection for the ISI channel with AWGN with inaccurate knowledge of the channel memory at the receiver. Without access to the accurate state transition matrix, the BCJR- NN performance significantly degrades in both settings. To this end, we devised an alternative approach for data-driven BCJR detection based on the unsupervised learning of a hidden Markov model (HMM). The BCJR-HMM allowed us to optimize both the likelihood function and the state transition matrix without labeling. Moreover, we demonstrated the viability of a hybrid NN and HMM BCJR detection where NN is used for learning the likelihoods, while the state transitions are optimized via HMM. While reducing the required prior channel knowledge, the examined data-driven detectors with learned trellis state transitions achieve bit error rates close to the optimal full CSI-based BCJR, significantly outperforming detection with inaccurate CSI.

cs.IT↗

arXiv2024

Analysis of Impulsive Interference in Digital Audio Broadcasting Systems in Electric Vehicles

Recently, new types of interference in electric vehicles (EVs), such as converters switching and/or battery chargers, have been found to degrade the performance of wireless digital transmission systems. Measurements show that such an interference is characterized by impulsive behavior and is widely varying in time. This paper uses recorded data from our EV testbed to analyze the impulsive interference in the digital audio broadcasting band. Moreover, we use our analysis to obtain a corresponding interference model. In particular, we studied the temporal characteristics of the interference and confirmed that its amplitude indeed exhibits an impulsive behavior. Our results show that impulsive events span successive received signal samples and thus indicate a bursty nature. To this end, we performed a data-driven modification of a well-established model for bursty impulsive interference, the Markov-Middleton model, to produce synthetic noise realization. We investigate the optimal symbol detector design based on the proposed model and show significant performance gains compared to the conventional detector based on the additive white Gaussian noise assumption.

eess.SP↗

arXiv2024

Europium $c$-axis ferromagnetism in Eu(Co$_{1-x}$Ni$_{x}$)$_{2-y}$As$_{2}$: A single-crystal neutron diffraction study

We report neutron diffraction results for the body-centered-tetragonal series Eu(Co$_{1-x}$Ni$_x$)$_{2-y}$As$_2$, $x=0.10$, $0.20$, $0.42$, and $0.82$, $y\leq0.10$, that detail changes to the magnetic ordering with nominal hole doping. We report the antiferromagnetic (AFM) propagation vectors, magnetic transition temperatures, and the ordered magnetic moments. We find a nonmonotonic change of the AFM propagation vector with $x$, with a minimum occurring at the tetragonal to collapsed-tetragonal phase crossover. For $x=0.10$ and $0.82$ we find $c$-axis helix ordering of the Eu magnetic moments (spins) similar to $x=0$ and $1$, with the spins oriented within the $ab$-plane. For $x=0.20$ and $0.42$ we find higher-temperature $c$-axis FM order and lower-temperature $c$-axis cone order. Using the extinction conditions for the space group, we discovered that the Eu spins are ordered in the higher-temperature $c$-axis FM phase for intermediate values of $x$, contrary to a previous report suggesting only Co/Ni spin ordering. Although we cannot directly confirm that the Co/Ni spins are also ordered, we suggest that $c$-axis itinerant-FM ordering of the Co/Ni spins could provide a molecular field that drives FM ordering of the Eu spins, which in turn provides the anisotropy for the lower-temperature $c$-axis cone order.

cond-mat.str-el↗

arXiv2024

Multi-perspective Feedback-attention Coupling Model for Continuous-time Dynamic Graphs

Recently, representation learning over graph networks has gained popularity, with various models showing promising results. Despite this, several challenges persist: 1) most methods are designed for static or discrete-time dynamic graphs; 2) existing continuous-time dynamic graph algorithms focus on a single evolving perspective; and 3) many continuous-time dynamic graph approaches necessitate numerous temporal neighbors to capture long-term dependencies. In response, this paper introduces the Multi-Perspective Feedback-Attention Coupling (MPFA) model. MPFA incorporates information from both evolving and raw perspectives, efficiently learning the interleaved dynamics of observed processes. The evolving perspective employs temporal self-attention to distinguish continuously evolving temporal neighbors for information aggregation. Through dynamic updates, this perspective can capture long-term dependencies using a small number of temporal neighbors. Meanwhile, the raw perspective utilizes a feedback attention module with growth characteristic coefficients to aggregate raw neighborhood information. Experimental results on a self-organizing dataset and seven public datasets validate the efficacy and competitiveness of our proposed model.

cs.LG↗

arXiv2024

CPA-Enhancer: Chain-of-Thought Prompted Adaptive Enhancer for Object Detection under Unknown Degradations

Object detection methods under known single degradations have been extensively investigated. However, existing approaches require prior knowledge of the degradation type and train a separate model for each, limiting their practical applications in unpredictable environments. To address this challenge, we propose a chain-of-thought (CoT) prompted adaptive enhancer, CPA-Enhancer, for object detection under unknown degradations. Specifically, CPA-Enhancer progressively adapts its enhancement strategy under the step-by-step guidance of CoT prompts, that encode degradation-related information. To the best of our knowledge, it's the first work that exploits CoT prompting for object detection tasks. Overall, CPA-Enhancer is a plug-and-play enhancement model that can be integrated into any generic detectors to achieve substantial gains on degraded images, without knowing the degradation type priorly. Experimental results demonstrate that CPA-Enhancer not only sets the new state of the art for object detection but also boosts the performance of other downstream vision tasks under unknown degradations.

cs.CV↗

arXiv2024

Protein Discovery with Discrete Walk-Jump Sampling

We resolve difficulties in training and sampling from a discrete generative model by learning a smoothed energy function, sampling from the smoothed data manifold with Langevin Markov chain Monte Carlo (MCMC), and projecting back to the true data manifold with one-step denoising. Our Discrete Walk-Jump Sampling formalism combines the contrastive divergence training of an energy-based model and improved sample quality of a score-based model, while simplifying training and sampling by requiring only a single noise level. We evaluate the robustness of our approach on generative modeling of antibody proteins and introduce the distributional conformity score to benchmark protein generative models. By optimizing and sampling from our models for the proposed distributional conformity score, 97-100% of generated samples are successfully expressed and purified and 70% of functional designs show equal or improved binding affinity compared to known functional antibodies on the first attempt in a single round of laboratory experiments. We also report the first demonstration of long-run fast-mixing MCMC chains where diverse antibody protein classes are visited in a single MCMC chain.

q-bio.BM↗

arXiv2024

AbDiffuser: Full-Atom Generation of in vitro Functioning Antibodies

We introduce AbDiffuser, an equivariant and physics-informed diffusion model for the joint generation of antibody 3D structures and sequences. AbDiffuser is built on top of a new representation of protein structure, relies on a novel architecture for aligned proteins, and utilizes strong diffusion priors to improve the denoising process. Our approach improves protein diffusion by taking advantage of domain knowledge and physics-based constraints; handles sequence-length changes; and reduces memory complexity by an order of magnitude, enabling backbone and side chain generation. We validate AbDiffuser in silico and in vitro. Numerical experiments showcase the ability of AbDiffuser to generate antibodies that closely track the sequence and structural properties of a reference set. Laboratory experiments confirm that all 16 HER2 antibodies discovered were expressed at high levels and that 57.1% of the selected designs were tight binders.

q-bio.BM↗

arXiv2023

Probing Three-Dimensional Magnetic Fields: II -- An Interpretable Convolutional Neural Network

Observing 3D magnetic fields, including orientation and strength, within the interstellar medium is vital but notoriously difficult. However, recent advances in our understanding of anisotropic magnetohydrodynamic (MHD) turbulence demonstrate that MHD turbulence and 3D magnetic fields leave their imprints on the intensity features of spectroscopic observations. Leveraging these theoretical frameworks, we propose a novel Convolutional Neural Network (CNN) model to extract this embedded information, enabling the probe of 3D magnetic fields. This model examines not only the plane-of-the-sky magnetic field orientation ($ϕ$), but also the magnetic field's inclination angle ($γ$) relative to the line-of-sight, and the total magnetization level (M$_A^{-1}$) of the cloud. We train the model using synthetic emission lines of $^{13}$CO (J = 1 - 0) and C$^{18}$O (J = 1 - 0), generated from 3D MHD simulations that span conditions from sub-Alfvénic to super-Alfvénic molecular clouds. Our tests confirm that the CNN model effectively reconstructs the 3D magnetic field topology and magnetization. The median uncertainties are under $5^\circ$ for both $ϕ$ and $γ$, and less than 0.2 for M$_A$ in sub-Alfvénic conditions (M$_A\approx0.5$). In super-Alfvénic scenarios (M$_A\approx2.0$), they are under $15^\circ$ for $ϕ$ and $γ$, and 1.5 for M$_A$. We applied this trained CNN model to the L1478 molecular cloud. Results show a strong agreement between the CNN-predicted magnetic field orientation and that derived from Planck 353 GHz polarization data. The CNN approach enabled us to construct the 3D magnetic field map for L1478, revealing a global inclination angle of $\approx76^\circ$ and a global M$_A$ of $\approx1.07$.

astro-ph.GA↗

arXiv2023

Visual-Policy Learning through Multi-Camera View to Single-Camera View Knowledge Distillation for Robot Manipulation Tasks

The use of multi-camera views simultaneously has been shown to improve the generalization capabilities and performance of visual policies. However, the hardware cost and design constraints in real-world scenarios can potentially make it challenging to use multiple cameras. In this study, we present a novel approach to enhance the generalization performance of vision-based Reinforcement Learning (RL) algorithms for robotic manipulation tasks. Our proposed method involves utilizing a technique known as knowledge distillation, in which a pre-trained ``teacher'' policy trained with multiple camera viewpoints guides a ``student'' policy in learning from a single camera viewpoint. To enhance the student policy's robustness against camera location perturbations, it is trained using data augmentation and extreme viewpoint changes. As a result, the student policy learns robust visual features that allow it to locate the object of interest accurately and consistently, regardless of the camera viewpoint. The efficacy and efficiency of the proposed method were evaluated both in simulation and real-world environments. The results demonstrate that the single-view visual student policy can successfully learn to grasp and lift a challenging object, which was not possible with a single-view policy alone. Furthermore, the student policy demonstrates zero-shot transfer capability, where it can successfully grasp and lift objects in real-world scenarios for unseen visual configurations.

cs.RO↗

arXiv2023

Dynamic Link Prediction for New Nodes in Temporal Graph Networks

Modelling temporal networks for dynamic link prediction of new nodes has many real-world applications, such as providing relevant item recommendations to new customers in recommender systems and suggesting appropriate posts to new users on social platforms. Unlike old nodes, new nodes have few historical links, which poses a challenge for the dynamic link prediction task. Most existing dynamic models treat all nodes equally and are not specialized for new nodes, resulting in suboptimal performances. In this paper, we consider dynamic link prediction of new nodes as a few-shot problem and propose a novel model based on the meta-learning principle to effectively mitigate this problem. Specifically, we develop a temporal encoder with a node-level span memory to obtain a new node embedding, and then we use a predictor to determine whether the new node generates a link. To overcome the few-shot challenge, we incorporate the encoder-predictor into the meta-learning paradigm, which can learn two types of implicit information during the formation of the temporal network through span adaptation and node adaptation. The acquired implicit information can serve as model initialisation and facilitate rapid adaptation to new nodes through a fine-tuning process on just a few links. Experiments on three publicly available datasets demonstrate the superior performance of our model compared to existing state-of-the-art methods.

cs.LG↗

arXiv2023

Classification of skyrmionic textures and extraction of Hamiltonian parameters via machine learning

Classifying skyrmionic textures and extracting magnetic Hamiltonian parameters are fundamental and demanding endeavors within the field of two-dimensional (2D) spintronics. By using micromagnetic simulation and machine learning (ML) methods, we theoretically realize the recognition of nine skyrmionic textures and the mining of magnetic Hamiltonian parameters from massive spin texture images in 2D Heisenberg model. For textures classification, a deep neural network (DNN) trained according to transfer learning is proposed to distinguish nine different skyrmionic textures. For parameters extraction, based on the textures generated by different Heisenberg exchange stiffness (J), Dzyaloshinskii-Moriya strength (D), and anisotropy constant (K), we apply a multi-input single-output (MISO) deep learning model (handling with both images and parameters) and a support vector regression (SVR) model (dealing with Fourier features) to extract the parameters embedded in the spin textures. The models for classification and extraction both achieve great results with the accuracy of 98% (DNN),90% (MISO) and 80% (SVR). Importantly, via our ML methods, the skyrmionic textures with blurred phase boundaries can be effectively distinguished, and the concluded formation conditions of various skyrmionic textures, especially the skyrmion crystal, are consistent with previous reports. Besides, our models demonstrate the mapping relationship between spin texture images and magnetic parameters, which proves the feasibility of extracting microscopic mechanisms from experimental images and has guiding significance for the experiments of spintronics.

cond-mat.mes-hall↗

arXiv2023

On a continuous time model of gradient descent dynamics and instability in deep learning

The recipe behind the success of deep learning has been the combination of neural networks and gradient-based optimization. Understanding the behavior of gradient descent however, and particularly its instability, has lagged behind its empirical success. To add to the theoretical tools available to study gradient descent we propose the principal flow (PF), a continuous time flow that approximates gradient descent dynamics. To our knowledge, the PF is the only continuous flow that captures the divergent and oscillatory behaviors of gradient descent, including escaping local minima and saddle points. Through its dependence on the eigendecomposition of the Hessian the PF sheds light on the recently observed edge of stability phenomena in deep learning. Using our new understanding of instability we propose a learning rate adaptation method which enables us to control the trade-off between training stability and test set evaluation performance.

stat.ML↗

arXiv2023

Investigating Vision Foundational Models for Tactile Representation Learning

Tactile representation learning (TRL) equips robots with the ability to leverage touch information, boosting performance in tasks such as environment perception and object manipulation. However, the heterogeneity of tactile sensors results in many sensor- and task-specific learning approaches. This limits the efficacy of existing tactile datasets, and the subsequent generalisability of any learning outcome. In this work, we investigate the applicability of vision foundational models to sensor-agnostic TRL, via a simple yet effective transformation technique to feed the heterogeneous sensor readouts into the model. Our approach recasts TRL as a computer vision (CV) problem, which permits the application of various CV techniques for tackling TRL-specific challenges. We evaluate our approach on multiple benchmark tasks, using datasets collected from four different tactile sensors. Empirically, we demonstrate significant improvements in task performance, model robustness, as well as cross-sensor and cross-task knowledge transferability with limited data requirements.

cs.RO↗

arXiv2023

Magnetic field effects on the quantum spin liquid behaviors of NaYbS$_2$

Spin-orbit coupling is an important ingredient to regulate the many-body physics, especially for many spin liquid candidate materials such as rare-earth magnets and Kitaev materials. The rare-earth chalcogenides NaYbCh$_2$ (Ch = O, S, Se) is a congenital frustrating system to exhibit the intrinsic landmark of spin liquid by eliminating both the site disorders between Na$^{+}$ and Yb$^{3+}$ ions with the big ionic size difference and the Dzyaloshinskii-Moriya interaction with the perfect triangular lattice of the Yb$^{3+}$ ions. The temperature versus magnetic-field phase diagram is established by the magnetization, specific heat, and neutron-scattering measurements. Notably, the neutron diffraction spectra and the magnetization curve might provide microscopic evidence for a series of spin configuration for in-plane fields, which include the disordered spin liquid state, 120$^{o}$ antiferromagnet, and one-half magnetization state. Furthermore, the ground state is suggested to be a gapless spin liquid from inelastic neutron scattering, and the magnetic field adjusts the spin orbit coupling. Therefore, the strong spin-orbit coupling in the frustrated quantum magnet substantially enriches low-energy spin physics. This rare-earth family could offer a good platform for exploring the quantum spin liquid ground state and quantum magnetic transitions.

cond-mat.str-el↗