arXiv subjects

Hao Chen

Publications and source records attributed to Hao Chen.

At least 1,261 records · Page 70Linked to original sources

arXiv2021

Algebraic Geometric Secret Sharing Schemes over Large Fields Are Asymptotically Threshold

In Chen-Cramer Crypto 2006 paper \cite{cc} algebraic geometric secret sharing schemes were proposed such that the "Fundamental Theorem in Information-Theoretically Secure Multiparty Computation" by Ben-Or, Goldwasser and Wigderson \cite{BGW88} and Chaum, Crépeau and Damgård \cite{CCD88} can be established over constant-size base finite fields. These algebraic geometric secret sharing schemes defined by a curve of genus $g$ over a constant size finite field ${\bf F}_q$ is quasi-threshold in the following sense, any subset of $u \leq T-1$ players (non qualified) has no information of the secret and any subset of $u \geq T+2g$ players (qualified) can reconstruct the secret. It is natural to ask that how far from the threshold these quasi-threshold secret sharing schemes are? How many subsets of $u \in [T, T+2g-1]$ players can recover the secret or have no information of the secret? In this paper it is proved that almost all subsets of $u \in [T,T+g-1]$ players have no information of the secret and almost all subsets of $u \in [T+g,T+2g-1]$ players can reconstruct the secret when the size $q$ goes to the infinity and the genus satisfies $\lim \frac{g}{\sqrt{q}}=0$. Then algebraic geometric secret sharing schemes over large finite fields are asymptotically threshold in this case. We also analyze the case when the size $q$ of the base field is fixed and the genus goes to the infinity.

cs.IT↗

arXiv2020

Learn molecular representations from large-scale unlabeled molecules for drug discovery

How to produce expressive molecular representations is a fundamental challenge in AI-driven drug discovery. Graph neural network (GNN) has emerged as a powerful technique for modeling molecular data. However, previous supervised approaches usually suffer from the scarcity of labeled data and have poor generalization capability. Here, we proposed a novel Molecular Pre-training Graph-based deep learning framework, named MPG, that leans molecular representations from large-scale unlabeled molecules. In MPG, we proposed a powerful MolGNet model and an effective self-supervised strategy for pre-training the model at both the node and graph-level. After pre-training on 11 million unlabeled molecules, we revealed that MolGNet can capture valuable chemistry insights to produce interpretable representation. The pre-trained MolGNet can be fine-tuned with just one additional output layer to create state-of-the-art models for a wide range of drug discovery tasks, including molecular properties prediction, drug-drug interaction, and drug-target interaction, involving 13 benchmark datasets. Our work demonstrates that MPG is promising to become a novel approach in the drug discovery pipeline.

cs.LG↗

arXiv2020

NUTA: Non-uniform Temporal Aggregation for Action Recognition

In the world of action recognition research, one primary focus has been on how to construct and train networks to model the spatial-temporal volume of an input video. These methods typically uniformly sample a segment of an input clip (along the temporal dimension). However, not all parts of a video are equally important to determine the action in the clip. In this work, we focus instead on learning where to extract features, so as to focus on the most informative parts of the video. We propose a method called the non-uniform temporal aggregation (NUTA), which aggregates features only from informative temporal segments. We also introduce a synchronization method that allows our NUTA features to be temporally aligned with traditional uniformly sampled video features, so that both local and clip-level features can be combined. Our model has achieved state-of-the-art performance on four widely used large-scale action-recognition datasets (Kinetics400, Kinetics700, Something-something V2 and Charades). In addition, we have created a visualization to illustrate how the proposed NUTA method selects only the most relevant parts of a video clip.

cs.CV↗

arXiv2020

Process monitoring based on orthogonal locality preserving projection with maximum likelihood estimation

By integrating two powerful methods of density reduction and intrinsic dimensionality estimation, a new data-driven method, referred to as OLPP-MLE (orthogonal locality preserving projection-maximum likelihood estimation), is introduced for process monitoring. OLPP is utilized for dimensionality reduction, which provides better locality preserving power than locality preserving projection. Then, the MLE is adopted to estimate intrinsic dimensionality of OLPP. Within the proposed OLPP-MLE, two new static measures for fault detection $T_{\scriptscriptstyle {OLPP}}^2$ and ${\rm SPE}_{\scriptscriptstyle {OLPP}}$ are defined. In order to reduce algorithm complexity and ignore data distribution, kernel density estimation is employed to compute thresholds for fault diagnosis. The effectiveness of the proposed method is demonstrated by three case studies.

stat.ME↗

arXiv2020

An improved mixture of probabilistic PCA for nonlinear data-driven process monitoring

An improved mixture of probabilistic principal component analysis (PPCA) has been introduced for nonlinear data-driven process monitoring in this paper. To realize this purpose, the technique of a mixture of probabilistic principal component analysers is utilized to establish the model of the underlying nonlinear process with local PPCA models, where a novel composite monitoring statistic is proposed based on the integration of two monitoring statistics in modified PPCA-based fault detection approach. Besides, the weighted mean of the monitoring statistics aforementioned is utilised as a metrics to detect potential abnormalities. The virtues of the proposed algorithm have been discussed in comparison with several unsupervised algorithms. Finally, Tennessee Eastman process and an autosuspension model are employed to demonstrate the effectiveness of the proposed scheme further.

stat.ME↗

arXiv2020

Importance of Electronic Correlation in the Intermetallic Half-Heusler Compounds

Low temperature scanning tunneling spectroscopy of HfNiSn shows a V^m(m < 1) zero bias anomaly around the Fermi level. This local density of states with a fractional power law shape is well known to be a consequence of electronic correlations. For comparison, we have also measured the tunneling conductances of other half-Heusler compounds with 18 valence electrons. ZrNiPb shows a metal-like local density of states, whereas ZrCoSb and NbFeSb show a linear and V^2 anomaly. One interpretation of these anomalies is that a correlation gap is opening in these compounds. By analyzing the magnetoresistance of HfNiSn, we demonstrate that at low temperatures, electron-electron scattering dominates. The T^m(m < 1) temperature dependence of the conductivity confirms that the electronic correlations are a bulk rather than a surface property.

cond-mat.dis-nn↗

arXiv2020

ByGARS: Byzantine SGD with Arbitrary Number of Attackers

We propose two novel stochastic gradient descent algorithms, ByGARS and ByGARS++, for distributed machine learning in the presence of any number of Byzantine adversaries. In these algorithms, reputation scores of workers are computed using an auxiliary dataset at the server. This reputation score is then used for aggregating the gradients for stochastic gradient descent. The computational complexity of ByGARS++ is the same as the usual distributed stochastic gradient descent method with only an additional inner product computation in every iteration. We show that using these reputation scores for gradient aggregation is robust to any number of multiplicative noise Byzantine adversaries and use two-timescale stochastic approximation theory to prove convergence for strongly convex loss functions. We demonstrate the effectiveness of the algorithms for non-convex learning problems using MNIST and CIFAR-10 datasets against almost all state-of-the-art Byzantine attacks. We also show that the proposed algorithms are robust to multiple different types of attacks at the same time.

cs.LG↗

arXiv2020

Backpropagating Linearly Improves Transferability of Adversarial Examples

The vulnerability of deep neural networks (DNNs) to adversarial examples has drawn great attention from the community. In this paper, we study the transferability of such examples, which lays the foundation of many black-box attacks on DNNs. We revisit a not so new but definitely noteworthy hypothesis of Goodfellow et al.'s and disclose that the transferability can be enhanced by improving the linearity of DNNs in an appropriate manner. We introduce linear backpropagation (LinBP), a method that performs backpropagation in a more linear fashion using off-the-shelf attacks that exploit gradients. More specifically, it calculates forward as normal but backpropagates loss as if some nonlinear activations are not encountered in the forward pass. Experimental results demonstrate that this simple yet effective method obviously outperforms current state-of-the-arts in crafting transferable adversarial examples on CIFAR-10 and ImageNet, leading to more effective attacks on a variety of DNNs.

cs.LG↗

arXiv2020

Practical No-box Adversarial Attacks against DNNs

The study of adversarial vulnerabilities of deep neural networks (DNNs) has progressed rapidly. Existing attacks require either internal access (to the architecture, parameters, or training set of the victim model) or external access (to query the model). However, both the access may be infeasible or expensive in many scenarios. We investigate no-box adversarial examples, where the attacker can neither access the model information or the training set nor query the model. Instead, the attacker can only gather a small number of examples from the same problem domain as that of the victim model. Such a stronger threat model greatly expands the applicability of adversarial attacks. We propose three mechanisms for training with a very small dataset (on the order of tens of examples) and find that prototypical reconstruction is the most effective. Our experiments show that adversarial examples crafted on prototypical auto-encoding models transfer well to a variety of image classification and face verification models. On a commercial celebrity recognition system held by clarifai.com, our approach significantly diminishes the average prediction accuracy of the system to only 15.40%, which is on par with the attack that transfers adversarial examples from a pre-trained Arcface model.

cs.CV↗

arXiv2020

BoxInst: High-Performance Instance Segmentation with Box Annotations

We present a high-performance method that can achieve mask-level instance segmentation with only bounding-box annotations for training. While this setting has been studied in the literature, here we show significantly stronger performance with a simple design (e.g., dramatically improving previous best reported mask AP of 21.1% in Hsu et al. (2019) to 31.6% on the COCO dataset). Our core idea is to redesign the loss of learning masks in instance segmentation, with no modification to the segmentation network itself. The new loss functions can supervise the mask training without relying on mask annotations. This is made possible with two loss terms, namely, 1) a surrogate term that minimizes the discrepancy between the projections of the ground-truth box and the predicted mask; 2) a pairwise loss that can exploit the prior that proximal pixels with similar colors are very likely to have the same category label. Experiments demonstrate that the redesigned mask loss can yield surprisingly high-quality instance masks with only box annotations. For example, without using any mask annotations, with a ResNet-101 backbone and 3x training schedule, we achieve 33.2% mask AP on COCO test-dev split (vs. 39.1% of the fully supervised counterpart). Our excellent experiment results on COCO and Pascal VOC indicate that our method dramatically narrows the performance gap between weakly and fully supervised instance segmentation. Code is available at: https://git.io/AdelaiDet

cs.CV↗

arXiv2020

Pareto Deterministic Policy Gradients and Its Application in 5G Massive MIMO Networks

In this paper, we consider jointly optimizing cell load balance and network throughput via a reinforcement learning (RL) approach, where inter-cell handover (i.e., user association assignment) and massive MIMO antenna tilting are configured as the RL policy to learn. Our rationale behind using RL is to circumvent the challenges of analytically modeling user mobility and network dynamics. To accomplish this joint optimization, we integrate vector rewards into the RL value network and conduct RL action via a separate policy network. We name this method as Pareto deterministic policy gradients (PDPG). It is an actor-critic, model-free and deterministic policy algorithm which can handle the coupling objectives with the following two merits: 1) It solves the optimization via leveraging the degree of freedom of vector reward as opposed to choosing handcrafted scalar-reward; 2) Cross-validation over multiple policies can be significantly reduced. Accordingly, the RL enabled network behaves in a self-organized way: It learns out the underlying user mobility through measurement history to proactively operate handover and antenna tilt without environment assumptions. Our numerical evaluation demonstrates that the introduced RL method outperforms scalar-reward based approaches. Meanwhile, to be self-contained, an ideal static optimization based brute-force search solver is included as a benchmark. The comparison shows that the RL approach performs as well as this ideal strategy, though the former one is constrained with limited environment observations and lower action frequency, whereas the latter ones have full access to the user mobility. The convergence of our introduced approach is also tested under different user mobility environment based on our measurement data from a real scenario.

cs.LG↗

arXiv2020

Enhancing Diversity in Teacher-Student Networks via Asymmetric branches for Unsupervised Person Re-identification

The objective of unsupervised person re-identification (Re-ID) is to learn discriminative features without labor-intensive identity annotations. State-of-the-art unsupervised Re-ID methods assign pseudo labels to unlabeled images in the target domain and learn from these noisy pseudo labels. Recently introduced Mean Teacher Model is a promising way to mitigate the label noise. However, during the training, self-ensembled teacher-student networks quickly converge to a consensus which leads to a local minimum. We explore the possibility of using an asymmetric structure inside neural network to address this problem. First, asymmetric branches are proposed to extract features in different manners, which enhances the feature diversity in appearance signatures. Then, our proposed cross-branch supervision allows one branch to get supervision from the other branch, which transfers distinct knowledge and enhances the weight diversity between teacher and student networks. Extensive experiments show that our proposed method can significantly surpass the performance of previous work on both unsupervised domain adaptation and fully unsupervised Re-ID tasks.

cs.CV↗

arXiv2020

Unifying Instance and Panoptic Segmentation with Dynamic Rank-1 Convolutions

Recently, fully-convolutional one-stage networks have shown superior performance comparing to two-stage frameworks for instance segmentation as typically they can generate higher-quality mask predictions with less computation. In addition, their simple design opens up new opportunities for joint multi-task learning. In this paper, we demonstrate that adding a single classification layer for semantic segmentation, fully-convolutional instance segmentation networks can achieve state-of-the-art panoptic segmentation quality. This is made possible by our novel dynamic rank-1 convolution (DR1Conv), a novel dynamic module that can efficiently merge high-level context information with low-level detailed features which is beneficial for both semantic and instance segmentation. Importantly, the proposed new method, termed DR1Mask, can perform panoptic segmentation by adding a single layer. To our knowledge, DR1Mask is the first panoptic segmentation framework that exploits a shared feature map for both instance and semantic segmentation by considering both efficacy and efficiency. Not only our framework is much more efficient -- twice as fast as previous best two-branch approaches, but also the unified framework opens up opportunities for using the same context module to improve the performance for both tasks. As a byproduct, when performing instance segmentation alone, DR1Mask is 10% faster and 1 point in mAP more accurate than previous state-of-the-art instance segmentation network BlendMask. Code is available at: https://git.io/AdelaiDet

cs.CV↗

arXiv2020

Three Point Energy Correlators in the Collinear Limit: Symmetries, Dualities and Analytic Results

Energy Correlators measure the energy deposited in multiple detectors as a function of the angles between the detectors. In this paper, we analytically compute the three particle correlator in the collinear limit in QCD for quark and gluon jets, and also in $\mathcal{N}=4$ super Yang-Mills theory. We find an intriguing duality between the integrals for the energy correlators and infrared finite Feynman parameter integrals, which maps the angles of the correlators to dual momentum variables. In $\mathcal{N}=4$, we use this duality to express our result as a rational sum of simple Feynman integrals (triangles and boxes). In QCD our result is expressed as a sum of the same transcendental functions, but with more complicated rational functions of cross ratio variables as coefficients. Our results represent the first analytic calculation of a three-prong jet substructure observable of phenomenological relevance for the LHC, revealing unexplored simplicity in the energy flow of QCD jets. They also provide valuable data for improving the understanding of the light-ray operator product expansion.

hep-ph↗

arXiv2020

Rethinking Jets with Energy Correlators: Tracks, Resummation and Analytic Continuation

We introduce an infinite set of jet substructure observables, derived as projections of $N$-point energy correlators, that are both convenient for experimental studies and maintain remarkable analytic properties derived from their representations in terms of a finite number of light ray operators. We show that these observables can be computed using tracking or charge information with a simple reweighting by integer moments of non-perturbative track or fragmentation functions. Our results for the projected $N$-point correlators are analytic functions of $N$, allowing us to derive resummed results to next-to-leading logarithmic accuracy for all $N$. We analytically continue our results to non-integer values of $N$ and define a corresponding analytic continuation of the observable, which we term a $ν$-correlator, that can be measured on jets of hadrons at the LHC. This enables observables that probe the leading twist collinear dynamics of jets to be placed into a single analytic family, which we hope will lead to new insights into jet substructure.

hep-ph↗

arXiv2020

Spectroscopic Evidence of Superconductivity Pairing at 83 K in Single-Layer FeSe/SrTiO3 Films

Single-layer FeSe films grown on the SrTiO3 substrate (FeSe/STO) have attracted much attention because of their possible record-high superconducting critical temperature Tc and distinct electronic structures in iron-based superconductors. However, it has been under debate on how high its Tc can really reach due to the inconsistency of the results obtained from the transport, magnetic and spectroscopic measurements. Here we report spectroscopic evidence of superconductivity pairing at 83 K in single-layer FeSe/STO films. By preparing high-quality single-layer FeSe/STO films, we observe for the first time strong superconductivity-induced Bogoliubov back-bending bands that extend to rather high binding energy ~100 meV by high-resolution angle-resolved photoemission measurements. The Bogoliubov back-bending band provides a new definitive benchmark of superconductivity pairing that is directly observed up to 83 K in the single-layer FeSe/STO films. Moreover, we find that the superconductivity pairing state can be further divided into two temperature regions of 64-83 K and below 64 K. We propose the 64-83 K region may be attributed to superconductivity fluctuation while the region below 64 K corresponds to the realization of long-range superconducting phase coherence. These results indicate that either Tc as high as 83 K is achievable in iron-based superconductors, or there is a pseudogap formation from superconductivity fluctuation in single-layer FeSe/STO films.

cond-mat.supr-con↗

arXiv2020

Stacking disorder in periodic minimal surfaces

We construct 1-parameter families of non-periodic embedded minimal surfaces of infinite genus in $T \times \mathbb{R}$, where $T$ denotes a flat 2-tori. Each of our families converges to a foliation of $T \times \mathbb{R}$ by $T$. These surfaces then lift to minimal surfaces in $\mathbb{R}^3$ that are periodic in horizontal directions but not periodic in the vertical direction. In the language of crystallography, our construction can be interpreted as disordered stacking of layers of periodically arranged catenoid necks. Our work is motivated by experimental observations of twinning defects in periodic minimal surfaces, which we reproduce as special cases of stacking disorder.

math.DG↗

arXiv2020

QarSUMO: A Parallel, Congestion-optimized Traffic Simulator

Traffic simulators are important tools for tasks such as urban planning and transportation management. Microscopic simulators allow per-vehicle movement simulation, but require longer simulation time. The simulation overhead is exacerbated when there is traffic congestion and most vehicles move slowly. This in particular hurts the productivity of emerging urban computing studies based on reinforcement learning, where traffic simulations are heavily and repeatedly used for designing policies to optimize traffic related tasks. In this paper, we develop QarSUMO, a parallel, congestion-optimized version of the popular SUMO open-source traffic simulator. QarSUMO performs high-level parallelization on top of SUMO, to utilize powerful multi-core servers and enables future extension to multi-node parallel simulation if necessary. The proposed design, while partly sacrificing speedup, makes QarSUMO compatible with future SUMO improvements. We further contribute such an improvement by modifying the SUMO simulation engine for congestion scenarios where the update computation of consecutive and slow-moving vehicles can be simplified. We evaluate QarSUMO with both real-world and synthetic road network and traffic data, and examine its execution time as well as simulation accuracy relative to the original, sequential SUMO.

cs.DC↗