arXiv subjects

Zhi Yang

Publications and source records attributed to Zhi Yang.

At least 91 records · Page 5Linked to original sources

arXiv2021

OpenBox: A Generalized Black-box Optimization Service

Black-box optimization (BBO) has a broad range of applications, including automatic machine learning, engineering, physics, and experimental design. However, it remains a challenge for users to apply BBO methods to their problems at hand with existing software packages, in terms of applicability, performance, and efficiency. In this paper, we build OpenBox, an open-source and general-purpose BBO service with improved usability. The modular design behind OpenBox also facilitates flexible abstraction and optimization of basic BBO components that are common in other existing systems. OpenBox is distributed, fault-tolerant, and scalable. To improve efficiency, OpenBox further utilizes "algorithm agnostic" parallelization and transfer learning. Our experimental results demonstrate the effectiveness and efficiency of OpenBox compared to existing systems.

cs.LG↗

arXiv2021

RIM: Reliable Influence-based Active Learning on Graphs

Message passing is the core of most graph models such as Graph Convolutional Network (GCN) and Label Propagation (LP), which usually require a large number of clean labeled data to smooth out the neighborhood over the graph. However, the labeling process can be tedious, costly, and error-prone in practice. In this paper, we propose to unify active learning (AL) and message passing towards minimizing labeling costs, e.g., making use of few and unreliable labels that can be obtained cheaply. We make two contributions towards that end. First, we open up a perspective by drawing a connection between AL enforcing message passing and social influence maximization, ensuring that the selected samples effectively improve the model performance. Second, we propose an extension to the influence model that incorporates an explicit quality factor to model label noise. In this way, we derive a fundamentally new AL selection criterion for GCN and LP--reliable influence maximization (RIM)--by considering quantity and quality of influence simultaneously. Empirical studies on public datasets show that RIM significantly outperforms current AL methods in terms of accuracy and efficiency.

cs.LG↗

arXiv2021

Node Dependent Local Smoothing for Scalable Graph Learning

Recent works reveal that feature or label smoothing lies at the core of Graph Neural Networks (GNNs). Concretely, they show feature smoothing combined with simple linear regression achieves comparable performance with the carefully designed GNNs, and a simple MLP model with label smoothing of its prediction can outperform the vanilla GCN. Though an interesting finding, smoothing has not been well understood, especially regarding how to control the extent of smoothness. Intuitively, too small or too large smoothing iterations may cause under-smoothing or over-smoothing and can lead to sub-optimal performance. Moreover, the extent of smoothness is node-specific, depending on its degree and local structure. To this end, we propose a novel algorithm called node-dependent local smoothing (NDLS), which aims to control the smoothness of every node by setting a node-specific smoothing iteration. Specifically, NDLS computes influence scores based on the adjacency matrix and selects the iteration number by setting a threshold on the scores. Once selected, the iteration number can be applied to both feature smoothing and label smoothing. Experimental results demonstrate that NDLS enjoys high accuracy -- state-of-the-art performance on node classifications tasks, flexibility -- can be incorporated with any models, scalability and efficiency -- can support large scale graphs with fast training.

cs.LG↗

arXiv2021

Hunting for the heavy quark spin symmetry partner of $Z_{cs}$

The discovery of a charged strange hidden-charm state $Z_{cs}$(3985) implies another higher $Z^*_{cs}$ state coupling to $\bar{D}_s^{*-}D^{*0} + c.c$ under the heavy quark spin symmetry. In this paper we discuss a possible hunt for it with data taken at existing facilities. We point out a hint for $Z^*_{cs}$ in the data of $\bar{B}_s^0 \to J/ψK^- K^+$ at LHCb, though weak, in line with the production mechanism of pentaquark $P_c$. We also study the triangular singularity which would possibly enhance the production of $Z^*_{cs}$ in electron-positron collision. Surprisingly, the production rate of $Z^*_{cs}$ is expected to be maximum at the $e^+ e^-$ center of mass energy of 4.648 GeV, which is lower than 4.681 GeV for $Z_{cs}$ due to the inverted coupling hierarchy of $D_{s1} \bar D K$ and $D_{s2} \bar D^* K$ in the triangle diagrams. Their bottom analogue under heavy quark flavor symmetry is also discussed. Our theoretical analysis would confront with future experiment of LHCb, BESIII, and Bell II.

hep-ph↗

arXiv2021

Semi-inclusive lepto-production of hidden-charm exotic hadrons

We investigate the semi-inclusive production of hidden-charm exotic states, including $X(3872)$, $Z_c$, $Z_{cs}$ and the pentaquark $P_c$ states, in lepton-proton scattering processes. These hadrons are close to the thresholds of a pair of charm and anticharm hadrons, and are assumed to have hadronic molecular structure as their main components. In order to give order-of-magnitude estimates of the cross sections, we use Pythia to simulate the short-distance productions of the constituent hadrons, which then rescatter to form the exotic hadrons. The estimates for the $X(3872)$ and $Z_c(3900)^\pm$ are not in conflict with the upper limits measured at the COMPASS experiment for the exclusive photoproduction process. The results here indicate that the considered hidden-charm states can be copiously produced at the proposed electron-ion colliders EicC and US-EIC.

hep-ph↗

arXiv2021

Cascade Bagging for Accuracy Prediction with Few Training Samples

Accuracy predictor is trained to predict the validation accuracy of an network from its architecture encoding. It can effectively assist in designing networks and improving Neural Architecture Search(NAS) efficiency. However, a high-performance predictor depends on adequate trainning samples, which requires unaffordable computation overhead. To alleviate this problem, we propose a novel framework to train an accuracy predictor under few training samples. The framework consists ofdata augmentation methods and an ensemble learning algorithm. The data augmentation methods calibrate weak labels and inject noise to feature space. The ensemble learning algorithm, termed cascade bagging, trains two-level models by sampling data and features. In the end, the advantages of above methods are proved in the Performance Prediciton Track of CVPR2021 1st Lightweight NAS Challenge. Our code is made public at: https://github.com/dlongry/Solutionto-CVPR2021-NAS-Track2.

cs.CV↗

arXiv2021

Improving Ranking Correlation of Supernet with Candidates Enhancement and Progressive Training

One-shot neural architecture search (NAS) applies weight-sharing supernet to reduce the unaffordable computation overhead of automated architecture designing. However, the weight-sharing technique worsens the ranking consistency of performance due to the interferences between different candidate networks. To address this issue, we propose a candidates enhancement method and progressive training pipeline to improve the ranking correlation of supernet. Specifically, we carefully redesign the sub-networks in the supernet and map the original supernet to a new one of high capacity. In addition, we gradually add narrow branches of supernet to reduce the degree of weight sharing which effectively alleviates the mutual interference between sub-networks. Finally, our method ranks the 1st place in the Supernet Track of CVPR2021 1st Lightweight NAS Challenge.

cs.CV↗

arXiv2021

Evaluating Deep Graph Neural Networks

Graph Neural Networks (GNNs) have already been widely applied in various graph mining tasks. However, they suffer from the shallow architecture issue, which is the key impediment that hinders the model performance improvement. Although several relevant approaches have been proposed, none of the existing studies provides an in-depth understanding of the root causes of performance degradation in deep GNNs. In this paper, we conduct the first systematic experimental evaluation to present the fundamental limitations of shallow architectures. Based on the experimental results, we answer the following two essential questions: (1) what actually leads to the compromised performance of deep GNNs; (2) when we need and how to build deep GNNs. The answers to the above questions provide empirical insights and guidelines for researchers to design deep and well-performed GNNs. To show the effectiveness of our proposed guidelines, we present Deep Graph Multi-Layer Perceptron (DGMLP), a powerful approach (a paradigm in its own right) that helps guide deep GNN designs. Experimental results demonstrate three advantages of DGMLP: 1) high accuracy -- it achieves state-of-the-art node classification performance on various datasets; 2) high flexibility -- it can flexibly choose different propagation and transformation depths according to graph size and sparsity; 3) high scalability and efficiency -- it supports fast training on large-scale graphs. Our code is available in https://github.com/zwt233/DGMLP.

cs.LG↗

arXiv2021

Grain: Improving Data Efficiency of Graph Neural Networks via Diversified Influence Maximization

Data selection methods, such as active learning and core-set selection, are useful tools for improving the data efficiency of deep learning models on large-scale datasets. However, recent deep learning models have moved forward from independent and identically distributed data to graph-structured data, such as social networks, e-commerce user-item graphs, and knowledge graphs. This evolution has led to the emergence of Graph Neural Networks (GNNs) that go beyond the models existing data selection methods are designed for. Therefore, we present Grain, an efficient framework that opens up a new perspective through connecting data selection in GNNs with social influence maximization. By exploiting the common patterns of GNNs, Grain introduces a novel feature propagation concept, a diversified influence maximization objective with novel influence and diversity functions, and a greedy algorithm with an approximation guarantee into a unified framework. Empirical studies on public datasets demonstrate that Grain significantly improves both the performance and efficiency of data selection (including active learning and core-set selection) for GNNs. To the best of our knowledge, this is the first attempt to bridge two largely parallel threads of research, data selection, and social influence maximization, in the setting of GNNs, paving new ways for improving data efficiency.

cs.LG↗

arXiv2021

ROD: Reception-aware Online Distillation for Sparse Graphs

Graph neural networks (GNNs) have been widely used in many graph-based tasks such as node classification, link prediction, and node clustering. However, GNNs gain their performance benefits mainly from performing the feature propagation and smoothing across the edges of the graph, thus requiring sufficient connectivity and label information for effective propagation. Unfortunately, many real-world networks are sparse in terms of both edges and labels, leading to sub-optimal performance of GNNs. Recent interest in this sparse problem has focused on the self-training approach, which expands supervised signals with pseudo labels. Nevertheless, the self-training approach inherently cannot realize the full potential of refining the learning performance on sparse graphs due to the unsatisfactory quality and quantity of pseudo labels. In this paper, we propose ROD, a novel reception-aware online knowledge distillation approach for sparse graph learning. We design three supervision signals for ROD: multi-scale reception-aware graph knowledge, task-based supervision, and rich distilled knowledge, allowing online knowledge transfer in a peer-teaching manner. To extract knowledge concealed in the multi-scale reception fields, ROD explicitly requires individual student models to preserve different levels of locality information. For a given task, each student would predict based on its reception-scale knowledge, while simultaneously a strong teacher is established on-the-fly by combining multi-scale knowledge. Our approach has been extensively evaluated on 9 datasets and a variety of graph-based tasks, including node classification, link prediction, and node clustering. The result demonstrates that ROD achieves state-of-art performance and is more robust for the graph sparsity.

cs.LG↗

arXiv2021

VolcanoML: Speeding up End-to-End AutoML via Scalable Search Space Decomposition

End-to-end AutoML has attracted intensive interests from both academia and industry, which automatically searches for ML pipelines in a space induced by feature engineering, algorithm/model selection, and hyper-parameter tuning. Existing AutoML systems, however, suffer from scalability issues when applying to application domains with large, high-dimensional search spaces. We present VolcanoML, a scalable and extensible framework that facilitates systematic exploration of large AutoML search spaces. VolcanoML introduces and implements basic building blocks that decompose a large search space into smaller ones, and allows users to utilize these building blocks to compose an execution plan for the AutoML problem at hand. VolcanoML further supports a Volcano-style execution model - akin to the one supported by modern database systems - to execute the plan constructed. Our evaluation demonstrates that, not only does VolcanoML raise the level of expressiveness for search space decomposition in AutoML, it also leads to actual findings of decomposition strategies that are significantly more efficient than the ones employed by state-of-the-art AutoML systems such as auto-sklearn.

cs.LG↗

arXiv2021

GMLP: Building Scalable and Flexible Graph Neural Networks with Feature-Message Passing

In recent studies, neural message passing has proved to be an effective way to design graph neural networks (GNNs), which have achieved state-of-the-art performance in many graph-based tasks. However, current neural-message passing architectures typically need to perform an expensive recursive neighborhood expansion in multiple rounds and consequently suffer from a scalability issue. Moreover, most existing neural-message passing schemes are inflexible since they are restricted to fixed-hop neighborhoods and insensitive to the actual demands of different nodes. We circumvent these limitations by a novel feature-message passing framework, called Graph Multi-layer Perceptron (GMLP), which separates the neural update from the message passing. With such separation, GMLP significantly improves the scalability and efficiency by performing the message passing procedure in a pre-compute manner, and is flexible and adaptive in leveraging node feature messages over various levels of localities. We further derive novel variants of scalable GNNs under this framework to achieve the best of both worlds in terms of performance and efficiency. We conduct extensive evaluations on 11 benchmark datasets, including large-scale datasets like ogbn-products and an industrial dataset, demonstrating that GMLP achieves not only the state-of-art performance, but also high training scalability and efficiency.

cs.LG↗

arXiv2021

A Portable, Self-Contained Neuroprosthetic Hand with Deep Learning-Based Finger Control

Objective: Deep learning-based neural decoders have emerged as the prominent approach to enable dexterous and intuitive control of neuroprosthetic hands. Yet few studies have materialized the use of deep learning in clinical settings due to its high computational requirements. Methods: Recent advancements of edge computing devices bring the potential to alleviate this problem. Here we present the implementation of a neuroprosthetic hand with embedded deep learning-based control. The neural decoder is designed based on the recurrent neural network (RNN) architecture and deployed on the NVIDIA Jetson Nano - a compacted yet powerful edge computing platform for deep learning inference. This enables the implementation of the neuroprosthetic hand as a portable and self-contained unit with real-time control of individual finger movements. Results: The proposed system is evaluated on a transradial amputee using peripheral nerve signals (ENG) with implanted intrafascicular microelectrodes. The experiment results demonstrate the system's capabilities of providing robust, high-accuracy (95-99%) and low-latency (50-120 msec) control of individual finger movements in various laboratory and real-world environments. Conclusion: Modern edge computing platforms enable the effective use of deep learning-based neural decoders for neuroprosthesis control as an autonomous system. Significance: This work helps pioneer the deployment of deep neural networks in clinical applications underlying a new class of wearable biomedical devices with embedded artificial intelligence.

cs.RO↗

arXiv2021

Strange molecular partners of the $Z_c(3900)$ and $Z_c(4020)$

Quantum Chromodynamics presents a series of exact and approximate symmetries which can be exploited to predict new hadrons from previously known ones. The $Z_c(3900)$ and $Z_c(4020)$ resonances, which have been theorized to be isovector $D^* \bar{D}$ and $D^* \bar{D}^*$ molecules [$I^G(J^{PC}) = 1^-(1^{+-})$], are no exception. Here we argue that from SU(3)-flavor symmetry, we should expect the existence of strange partners of the $Z_c$'s with hadronic molecular configurations $D^* \bar{D}_s$-$D \bar{D}_s^*$ and $D^* \bar{D}^{*}_s$ (or, equivalently, quark content $c\bar{c} s \bar{q}$, with $q = u, d$). The quantum numbers of these $Z_{cs}$ and $Z_{cs}^*$ resonances would be $I(J^P)$ = $\frac{1}{2}(1^+)$. The predicted masses of these partners depend on the details of the theoretical scheme used, but they should be around the $D^* \bar{D}_s$-$D \bar{D}_s^*$ and $D^* \bar{D}^{*}_s$ thresholds, respectively. Moreover, any of these states could be either a virtual pole or a resonance. We show that, together with a possible triangle singularity contribution, such a picture nicely agrees with the very recent BESIII data of the $e^+e^-\to K^+(D_s^-D^{*0}+D_s^{*-}D^0)$.

hep-ph↗

arXiv2021

Photoproduction of strange hidden-charm and hidden-bottom states

Recently BESIII collaboration discovered a charged strange hidden-charm state $Z_{cs}$(3985) in the $D_s^-D^{*0} + D_s^{*-}D^{0}$ spectrum. A higher $Z'_{cs}$ state coupling to $\bar{D}_s^{*-}D^{*0}$ is expected by SU(3)-flavor symmetry, and their bottom partners are anticipated by heavy quark flavor symmetry. Here we study the photoproduction of these exotic states and investigate carefully the background from Pomeron exchange. Our results indicate that the maximal photoproduction cross section of strange partner is around 1 $\sim$ 2 orders of magnitude smaller than that of the corresponding non-strange states. The possibility of searching for them in future electron-ion colliders (EIC) is briefly discussed.

hep-ph↗

arXiv2021

Electron-Ion Collider in China

Lepton scattering is an established ideal tool for studying inner structure of small particles such as nucleons as well as nuclei. As a future high energy nuclear physics project, an Electron-ion collider in China (EicC) has been proposed. It will be constructed based on an upgraded heavy-ion accelerator, High Intensity heavy-ion Accelerator Facility (HIAF) which is currently under construction, together with a new electron ring. The proposed collider will provide highly polarized electrons (with a polarization of $\sim$80%) and protons (with a polarization of $\sim$70%) with variable center of mass energies from 15 to 20 GeV and the luminosity of (2-3) $\times$ 10$^{33}$ cm$^{-2}$ s$^{-1}$. Polarized deuterons and Helium-3, as well as unpolarized ion beams from Carbon to Uranium, will be also available at the EicC. The main foci of the EicC will be precision measurements of the structure of the nucleon in the sea quark region, including 3D tomography of nucleon; the partonic structure of nuclei and the parton interaction with the nuclear environment; the exotic states, especially those with heavy flavor quark contents. In addition, issues fundamental to understanding the origin of mass could be addressed by measurements of heavy quarkonia near-threshold production at the EicC. In order to achieve the above-mentioned physics goals, a hermetical detector system will be constructed with cutting-edge technologies. This document is the result of collective contributions and valuable inputs from experts across the globe. The EicC physics program complements the ongoing scientific programs at the Jefferson Laboratory and the future EIC project in the United States. The success of this project will also advance both nuclear and particle physics as well as accelerator and detector technology in China.

nucl-ex↗

arXiv2020

Mechanical Properties of Atomically Thin Boron Nitride and the Role of Interlayer Interactions

Atomically thin boron nitride (BN) nanosheets are important two-dimensional nanomaterials with many unique properties distinct from those of graphene, but the investigation of their mechanical properties still greatly lacks. Here we report that high-quality single-crystalline mono- and few-layer BN nanosheets are one of the strongest electrically insulating materials. More intriguingly, few-layer BN shows mechanical behaviors quite different from those of few-layer graphene under indentation. In striking contrast to graphene, whose strength decreases by more than 30% when the number of layers increases from 1 to 8, the mechanical strength of BN nanosheets is not sensitive to increasing thickness. We attribute this difference to the distinct interlayer interactions and hence sliding tendencies in these two materials under indentation. The significantly better mechanical integrity of BN nanosheets makes them a more attractive candidate than graphene for several applications, e.g. as mechanical reinforcements.

cond-mat.mtrl-sci↗

arXiv2020

Fine-Grained Image Captioning with Global-Local Discriminative Objective

Significant progress has been made in recent years in image captioning, an active topic in the fields of vision and language. However, existing methods tend to yield overly general captions and consist of some of the most frequent words/phrases, resulting in inaccurate and indistinguishable descriptions (see Figure 1). This is primarily due to (i) the conservative characteristic of traditional training objectives that drives the model to generate correct but hardly discriminative captions for similar images and (ii) the uneven word distribution of the ground-truth captions, which encourages generating highly frequent words/phrases while suppressing the less frequent but more concrete ones. In this work, we propose a novel global-local discriminative objective that is formulated on top of a reference model to facilitate generating fine-grained descriptive captions. Specifically, from a global perspective, we design a novel global discriminative constraint that pulls the generated sentence to better discern the corresponding image from all others in the entire dataset. From the local perspective, a local discriminative constraint is proposed to increase attention such that it emphasizes the less frequent but more concrete words/phrases, thus facilitating the generation of captions that better describe the visual details of the given images. We evaluate the proposed method on the widely used MS-COCO dataset, where it outperforms the baseline methods by a sizable margin and achieves competitive performance over existing leading approaches. We also conduct self-retrieval experiments to demonstrate the discriminability of the proposed method.

cs.CV↗