Searcharxiv⌕ Search

arXiv subjects

Yuan Meng

Publications and source records attributed to Yuan Meng.

At least 73 records · Page 4Linked to original sources
arXiv2024

Optical skyrmions from metafibers

Optical skyrmions are an emerging class of structured light with sophisticated particle-like topologies with great potential for revolutionizing modern informatics. However, the current generation of optical skyrmions involves complex or bulky systems, hindering their development of practical applications. Here, exploiting the emergent "lab-on-fiber" technology, we demonstrate the design of a metafiber-integrated photonic skyrmion generator. We not only successfully generated high-quality optical skyrmions from metafibers, but also experimentally verified their remarkable properties, such as regulability and topological stability with deep-subwavelength features beyond the diffraction limits. Our flexible and fiber-integrated optical skyrmions platform paves the avenue for future applications of topologically-enhanced remote super-resolution microscopy and super-robust information transfer.

physics.optics↗
arXiv2024

TMPQ-DM: Joint Timestep Reduction and Quantization Precision Selection for Efficient Diffusion Models

Diffusion models have emerged as preeminent contenders in the realm of generative models. Distinguished by their distinctive sequential generative processes, characterized by hundreds or even thousands of timesteps, diffusion models progressively reconstruct images from pure Gaussian noise, with each timestep necessitating full inference of the entire model. However, the substantial computational demands inherent to these models present challenges for deployment, quantization is thus widely used to lower the bit-width for reducing the storage and computing overheads. Current quantization methodologies primarily focus on model-side optimization, disregarding the temporal dimension, such as the length of the timestep sequence, thereby allowing redundant timesteps to continue consuming computational resources, leaving substantial scope for accelerating the generative process. In this paper, we introduce TMPQ-DM, which jointly optimizes timestep reduction and quantization to achieve a superior performance-efficiency trade-off, addressing both temporal and model optimization aspects. For timestep reduction, we devise a non-uniform grouping scheme tailored to the non-uniform nature of the denoising process, thereby mitigating the explosive combinations of timesteps. In terms of quantization, we adopt a fine-grained layer-wise approach to allocate varying bit-widths to different layers based on their respective contributions to the final generative performance, thus rectifying performance degradation observed in prior studies. To expedite the evaluation of fine-grained quantization, we further devise a super-network to serve as a precision solver by leveraging shared quantization results. These two design components are seamlessly integrated within our framework, enabling rapid joint exploration of the exponentially large decision space via a gradient-free evolutionary search algorithm.

cs.CV↗
arXiv2024

Investigating the Impact of Quantization on Adversarial Robustness

Quantization is a promising technique for reducing the bit-width of deep models to improve their runtime performance and storage efficiency, and thus becomes a fundamental step for deployment. In real-world scenarios, quantized models are often faced with adversarial attacks which cause the model to make incorrect inferences by introducing slight perturbations. However, recent studies have paid less attention to the impact of quantization on the model robustness. More surprisingly, existing studies on this topic even present inconsistent conclusions, which prompted our in-depth investigation. In this paper, we conduct a first-time analysis of the impact of the quantization pipeline components that can incorporate robust optimization under the settings of Post-Training Quantization and Quantization-Aware Training. Through our detailed analysis, we discovered that this inconsistency arises from the use of different pipelines in different studies, specifically regarding whether robust optimization is performed and at which quantization stage it occurs. Our research findings contribute insights into deploying more secure and robust quantized networks, assisting practitioners in reference for scenarios with high-security requirements and limited resources.

cs.LG↗
arXiv2024

Images of hairy Reissner-Nordström black hole illuminated by static accretions

We investigate the shadow and optical appearance of the hairy Reissner-Nordström (RN) black hole illuminated by two toy models of static accretion. The hairy RN black hole was constructed in the gravitation decoupling approach to describe the deformation of a Schwarzschild black hole due to the inclusion of additional arbitrary source (scalar field, tensor field, fluidlike dark matter, etc). So it is characterized by the parameters: mass ($M$), deformation factor ($α$), electric charge ($Q$) and the additional hairy charge ($l_o$), differentiating from the case in RN black hole. Though the specific background theory that results in this hairy RN black hole is still tricky, here we shall focus on the novel observable features introduced by the hair of this black hole. First, we find that for the hairy RN black hole, the event horizon, radius of photon sphere and critical impact parameter all increase as the increasings of $Q$ and $l_o$, but decrease as $α$ grows. The three characterized parameters are found to have significant effects on the photon trajectories, shadows and images of hairy RN black hole surrounded by the static accretion disk and spherical accretion, respectively. Particularly, both $Q$ and $l_o$ have mutually reinforcing effects on the optical appearance and shadows of the hairy RN black hole, which implies that we may not distinguish the electric charge and hairy charge from the shadow and image of black hole in this scenario. Additionally, because of the competing effects of the charge parameters ($Q, l_o$) and the deviation parameter $α$ on the observed intensities of brightness, the optical appearance between the hairy RN black hole and RN black hole could have degeneracies, indicating the indistinguishability. Our current results contribute more to the phenomenal aspects which could be helpful to build the background theory of hairy RN black hole.

gr-qc↗
arXiv2023

Single-image based deep learning for precise atomic defects identification

Defect engineering has been profoundly employed to confer desirable functionality to materials that pristine lattices inherently lack. Although single atomic-resolution scanning transmission electron microscopy (STEM) images are widely accessible for defect engineering, harnessing atomic-scale images containing various defects through traditional image analysis methods is hindered by random noise and human bias. Yet the rise of deep learning (DL) offering an alternative approach, its widespread application is primarily restricted by the need for large amounts of training data with labeled ground truth. In this study, we propose a two-stage method to address the problems of high annotation cost and image noise in the detection of atomic defects in monolayer 2D materials. In the first stage, to tackle the issue of data scarcity, we employ a two-state transformation network based on U-GAT-IT for adding realistic noise to simulated images with pre-located ground truth labels, thereby infinitely expanding the training dataset. In the second stage, atomic defects in monolayer 2D materials are effectively detected with high accuracy using U-Net models trained with the data generated in the first stage, avoiding random noise and human bias issues. In both stages, we utilize segmented unit-cell-level images to simplify the model's task and enhance its accuracy. Our results demonstrate that not only sulfur vacancies, we are also able to visualize oxygen dopants in monolayer MoS2, which are usually overwhelmed by random background noise. As the training was based on a few segmented unit-cell-level realistic images, this method can be readily extended to other 2D materials. Therefore, our results outline novel ways to train the model with minimized datasets, offering great opportunities to fully exploit the power of machine learning (ML) applicable to a broad materials science community.

cond-mat.mtrl-sci↗
arXiv2023

DIVA: A Dirichlet Process Mixtures Based Incremental Deep Clustering Algorithm via Variational Auto-Encoder

Generative model-based deep clustering frameworks excel in classifying complex data, but are limited in handling dynamic and complex features because they require prior knowledge of the number of clusters. In this paper, we propose a nonparametric deep clustering framework that employs an infinite mixture of Gaussians as a prior. Our framework utilizes a memoized online variational inference method that enables the "birth" and "merge" moves of clusters, allowing our framework to cluster data in a "dynamic-adaptive" manner, without requiring prior knowledge of the number of features. We name the framework as DIVA, a Dirichlet Process-based Incremental deep clustering framework via Variational Auto-Encoder. Our framework, which outperforms state-of-the-art baselines, exhibits superior performance in classifying complex data with dynamically changing features, particularly in the case of incremental features. We released our source code implementation at: https://github.com/Ghiara/diva

cs.LG↗
arXiv2023

A Software-Hardware Co-Optimized Toolkit for Deep Reinforcement Learning on Heterogeneous Platforms

Deep Reinforcement Learning (DRL) is vital in various AI applications. DRL algorithms comprise diverse compute kernels, which may not be simultaneously optimized using a homogeneous architecture. However, even with available heterogeneous architectures, optimizing DRL performance remains a challenge due to the complexity of hardware and programming models employed in modern data centers. To address this, we introduce PEARL, a toolkit for composing parallel DRL systems on heterogeneous platforms consisting of general-purpose processors (CPUs) and accelerators (GPUs, FPGAs). Our innovations include: 1. A general training protocol agnostic of the underlying hardware, enabling portable implementations across various processors and accelerators. 2. Incorporation of DRL-specific scheduling optimizations within the protocol, facilitating parallelized training and enhancing the overall system performance. 3. High-level API for productive development using the toolkit. 4. Automatic optimization of DRL task-to-device assignments through performance estimation, supporting various optimization metrics including throughput and power efficiency. We showcase our toolkit through experimentation with two widely used DRL algorithms, DQN and DDPG, on two diverse heterogeneous platforms. The generated implementations outperform state-of-the-art libraries for CPU-GPU platforms by throughput improvements of up to 2.1$\times$ and power efficiency improvements of up to 3.4$\times$.

cs.DC↗
arXiv2023

Towards Fair and Comprehensive Comparisons for Image-Based 3D Object Detection

In this work, we build a modular-designed codebase, formulate strong training recipes, design an error diagnosis toolbox, and discuss current methods for image-based 3D object detection. In particular, different from other highly mature tasks, e.g., 2D object detection, the community of image-based 3D object detection is still evolving, where methods often adopt different training recipes and tricks resulting in unfair evaluations and comparisons. What is worse, these tricks may overwhelm their proposed designs in performance, even leading to wrong conclusions. To address this issue, we build a module-designed codebase and formulate unified training standards for the community. Furthermore, we also design an error diagnosis toolbox to measure the detailed characterization of detection models. Using these tools, we analyze current methods in-depth under varying settings and provide discussions for some open questions, e.g., discrepancies in conclusions on KITTI-3D and nuScenes datasets, which have led to different dominant methods for these datasets. We hope that this work will facilitate future research in image-based 3D object detection. Our codes will be released at \url{https://github.com/OpenGVLab/3dodi}

cs.CV↗
arXiv2023

Accelerating Deep Neural Network guided MCTS using Adaptive Parallelism

Deep Neural Network guided Monte-Carlo Tree Search (DNN-MCTS) is a powerful class of AI algorithms. In DNN-MCTS, a Deep Neural Network model is trained collaboratively with a dynamic Monte-Carlo search tree to guide the agent towards actions that yields the highest returns. While the DNN operations are highly parallelizable, the search tree operations involved in MCTS are sequential and often become the system bottleneck. Existing MCTS parallel schemes on shared-memory multi-core CPU platforms either exploit data parallelism but sacrifice memory access latency, or take advantage of local cache for low-latency memory accesses but constrain the tree search to a single thread. In this work, we analyze the tradeoff of these parallel schemes and develop performance models for both parallel schemes based on the application and hardware parameters. We propose a novel implementation that addresses the tradeoff by adaptively choosing the optimal parallel scheme for the MCTS component on the CPU. Furthermore, we propose an efficient method for searching the optimal communication batch size as the MCTS component on the CPU interfaces with DNN operations offloaded to an accelerator (GPU). Using a representative DNN-MCTS algorithm - Alphazero on board game benchmarks, we show that the parallel framework is able to adaptively generate the best-performing parallel implementation, leading to a range of $1.5\times - 3\times$ speedup compared with the baseline methods on CPU and CPU-GPU platforms.

cs.PF↗
arXiv2023

Characterizing Speed Performance of Multi-Agent Reinforcement Learning

Multi-Agent Reinforcement Learning (MARL) has achieved significant success in large-scale AI systems and big-data applications such as smart grids, surveillance, etc. Existing advancements in MARL algorithms focus on improving the rewards obtained by introducing various mechanisms for inter-agent cooperation. However, these optimizations are usually compute- and memory-intensive, thus leading to suboptimal speed performance in end-to-end training time. In this work, we analyze the speed performance (i.e., latency-bounded throughput) as the key metric in MARL implementations. Specifically, we first introduce a taxonomy of MARL algorithms from an acceleration perspective categorized by (1) training scheme and (2) communication method. Using our taxonomy, we identify three state-of-the-art MARL algorithms - Multi-Agent Deep Deterministic Policy Gradient (MADDPG), Target-oriented Multi-agent Communication and Cooperation (ToM2C), and Networked Multi-Agent RL (NeurComm) - as target benchmark algorithms, and provide a systematic analysis of their performance bottlenecks on a homogeneous multi-core CPU platform. We justify the need for MARL latency-bounded throughput to be a key performance metric in future literature while also addressing opportunities for parallelization and acceleration.

cs.LG↗
arXiv2023

RGAT: A Deeper Look into Syntactic Dependency Information for Coreference Resolution

Although syntactic information is beneficial for many NLP tasks, combining it with contextual information between words to solve the coreference resolution problem needs to be further explored. In this paper, we propose an end-to-end parser that combines pre-trained BERT with a Syntactic Relation Graph Attention Network (RGAT) to take a deeper look into the role of syntactic dependency information for the coreference resolution task. In particular, the RGAT model is first proposed, then used to understand the syntactic dependency graph and learn better task-specific syntactic embeddings. An integrated architecture incorporating BERT embeddings and syntactic embeddings is constructed to generate blending representations for the downstream task. Our experiments on a public Gendered Ambiguous Pronouns (GAP) dataset show that with the supervision learning of the syntactic dependency graph and without fine-tuning the entire BERT, we increased the F1-score of the previous best model (RGCN-with-BERT) from 80.3% to 82.5%, compared to the F1-score by single BERT embeddings from 78.5% to 82.5%. Experimental results on another public dataset - OntoNotes 5.0 demonstrate that the performance of the model is also improved by incorporating syntactic dependency information learned from RGAT.

cs.CL↗
arXiv2023

SEAM: Searching Transferable Mixed-Precision Quantization Policy through Large Margin Regularization

Mixed-precision quantization (MPQ) suffers from the time-consuming process of searching the optimal bit-width allocation i.e., the policy) for each layer, especially when using large-scale datasets such as ISLVRC-2012. This limits the practicality of MPQ in real-world deployment scenarios. To address this issue, this paper proposes a novel method for efficiently searching for effective MPQ policies using a small proxy dataset instead of the large-scale dataset used for training the model. Deviating from the established norm of employing a consistent dataset for both model training and MPQ policy search stages, our approach, therefore, yields a substantial enhancement in the efficiency of MPQ exploration. Nonetheless, using discrepant datasets poses challenges in searching for a transferable MPQ policy. Driven by the observation that quantization noise of sub-optimal policy exerts a detrimental influence on the discriminability of feature representations -- manifesting as diminished class margins and ambiguous decision boundaries -- our method aims to identify policies that uphold the discriminative nature of feature representations, i.e., intra-class compactness and inter-class separation. This general and dataset-independent property makes us search for the MPQ policy over a rather small-scale proxy dataset and then the policy can be directly used to quantize the model trained on a large-scale dataset. Our method offers several advantages, including high proxy data utilization, no excessive hyper-parameter tuning, and high searching efficiency. We search high-quality MPQ policies with the proxy dataset that has only 4% of the data scale compared to the large-scale target dataset, achieving the same accuracy as searching directly on the latter, improving MPQ searching efficiency by up to 300 times.

cs.CV↗
arXiv2023

Images from disk and spherical accretions of hairy Schwarzschild black holes

A hairy Schwarzschild black hole describes the deformation of Schwarzschild black hole due to including additional sources. It is found that depending on the hairy parameters, the photons' configurations around this black hole can be classified into two cases, corresponding to the hairy Schwarzschild black hole with single photon sphere and double photon spheres, respectively. We focus on the shadows and images of the hairy Schwarzschild black hole under two types of static thin illuminations conditions: disk accretion and spherical accretion, respectively. Under both illuminations, the two hairy parameters ($α$ and $l_o$) have competitive effects on the shadow and optical appearance image of the hairy Schwarzschild black hole with single photon sphere. This means that even though the parameters have significant influences on the rings and shadows, its images with certain groups of $α$ and $l_o$ could be indistinguishable to that of Schwarzschild black hole, namely, the degeneracies of images raise between the hairy Schwarzschild black hole and Schwarzschild black hole. Moreover, the optical appearance image of the hairy Schwarzschild black hole with double photon spheres will exhibit new additional rings and accretion features, which {are not} present in the images of (hairy) Schwarzschild black hole with single photon sphere. Our theoretical studies on the rings and shadows provide a potential tool to differentiate the hairy Schwarzschild black hole with double photon spheres from Schwarzschild black hole, but they are not always helpful for the cases with single photon sphere due to the degeneracies.

gr-qc↗
arXiv2023

Contact-aware Shaping and Maintenance of Deformable Linear Objects With Fixtures

Studying the manipulation of deformable linear objects has significant practical applications in industry, including car manufacturing, textile production, and electronics automation. However, deformable linear object manipulation poses a significant challenge in developing planning and control algorithms, due to the precise and continuous control required to effectively manipulate the deformable nature of these objects. In this paper, we propose a new framework to control and maintain the shape of deformable linear objects with two robot manipulators utilizing environmental contacts. The framework is composed of a shape planning algorithm which automatically generates appropriate positions to place fixtures, and an object-centered skill engine which includes task and motion planning to control the motion and force of both robots based on the object status. The status of the deformable linear object is estimated online utilizing visual as well as force information. The framework manages to handle a cable routing task in real-world experiments with two Panda robots and especially achieves contact-aware and flexible clip fixing with challenging fixtures.

cs.RO↗
arXiv2023

Rings and images of Horndeski hairy black hole illuminated by various thin accretions

We analyze the light rays around a static hairy black hole in Horndeski gravity with the use of ray-tracing procedure. We find that a stronger Horndeski hairy parameter corresponds to larger photon sphere as well as critical impact parameter, and wider ranges of photon ring and lensed ring emissions. These influences can be robustly interpreted from the shape of the effective potential of the photon's radial motion. Based on the distribution of the light rays, we then investigate the optical appearances of the Horndeski hairy black hole surrounded by various thin accretions. Firstly, we consider that the Horndeski hairy black hole is illuminated by the optically and geometrically thin accretion disk. We carefully clarify the contributions from the direct, lensed ring and photon ring intensities to the total observed intensity via the transfer function, which is rarely discussed in this scenario. We find that the Horndeski hair has significant influences on both shadow size and distributions of direct, lensed ring and photon ring brightness in three standard emission profiles. As a result, the rings and images of Horndeski hairy black hole and the origination of their brightness differentiate from those of Schwarzschild black hole (SBH). Then, when the Horndeski hairy black hole is illuminated by thin spherical accretions, the hairy black hole's shadow surrounded by a bright ring is larger than that of SBH, but the brightness of ring is fainter. Similar to that of SBH, the size of hairy black hole shadow does not change as the radial moving of the spherical accretion, and the brightness for the infalling accretion is fainter than that for the static accretion due to the Doppler effect. Therefore, we argue that the black hole image consisting of the shadow and accretion construction could, in theory, reflect the observational differences between the Horndeski hairy black hole and SBH.

gr-qc↗
arXiv2023

Allying nanophotonic structures with two-dimensional van der Waals materials

The integration of two-dimensional (2D) materials with photonic structures has catalyzed a wide spectrum of optical and optoelectronic applications. Conventional nanophotonic structures generally lack efficient reconfigurability and multifunctionality. The atomically thin 2D van der Waals materials can thus infuse new functionality and reconfigurability to the well-established library of photonic structures such as integrated waveguides, optical fibers, photonic crystals, micro-cavities, and metasurface, to name a few. Thanks to the handiness of van der Waals interfaces, the 2D materials can be easily transferred and mixed with other prefabricated photonic templates with high degrees of freedom, and can act as the optical gain, modulation, sensing, or plasmonic media for diverse applications. Here we review recent advents on combining 2D materials to nanophotonic structures for new functionality development or performance enhancements. Challenges and emerging opportunities in integrating van der Waals building blocks beyond 2D materials are also discussed.

physics.optics↗
arXiv2023

Shadow revisiting and weak gravitational lensing with Chern-Simons modification

Dynamical Chern-Simons (dCS) gravity has been attracting plenty of attentions due to the fact that it is a parity-violating modified theory of gravity that corresponds to a well-posed effective field theory in weak coupling approximation. In particular, a rotating black hole in dCS gravity is in contrast to the general relativistic counterparts. In this paper, we revisit the shadow of analytical rotating black hole spacetime in dCS modified gravity, based on which we study the shadow observables, discuss the constraint on the model parameters from the Event Horizon Telescope (EHT) observations, and analyze the real part of quasi-normal modes (QNMs) in the eikonal limit. In addition, we explore the deflection angle in weak gravitational field limit with the use of Gauss-Bonnet theorem. We find that the shadow related physics and the weak gravitational lensing effect are significantly influenced by the CS coupling, which could provide theoretical predictions for a future test of the dCS theory with EHT observations.

gr-qc↗
arXiv2023

DNG: Taxonomy Expansion by Exploring the Intrinsic Directed Structure on Non-gaussian Space

Taxonomy expansion is the process of incorporating a large number of additional nodes (i.e., "queries") into an existing taxonomy (i.e., "seed"), with the most important step being the selection of appropriate positions for each query. Enormous efforts have been made by exploring the seed's structure. However, existing approaches are deficient in their mining of structural information in two ways: poor modeling of the hierarchical semantics and failure to capture directionality of is-a relation. This paper seeks to address these issues by explicitly denoting each node as the combination of inherited feature (i.e., structural part) and incremental feature (i.e., supplementary part). Specifically, the inherited feature originates from "parent" nodes and is weighted by an inheritance factor. With this node representation, the hierarchy of semantics in taxonomies (i.e., the inheritance and accumulation of features from "parent" to "child") could be embodied. Additionally, based on this representation, the directionality of is-a relation could be easily translated into the irreversible inheritance of features. Inspired by the Darmois-Skitovich Theorem, we implement this irreversibility by a non-Gaussian constraint on the supplementary feature. A log-likelihood learning objective is further utilized to optimize the proposed model (dubbed DNG), whereby the required non-Gaussianity is also theoretically ensured. Extensive experimental results on two real-world datasets verify the superiority of DNG relative to several strong baselines.

cs.AI↗