arXiv subjects

Jian Sun

Publications and source records attributed to Jian Sun.

At least 325 records · Page 18Linked to original sources

arXiv2021

Resilient Control under Quantization and Denial-of-Service: Co-designing a Deadbeat Controller and Transmission Protocol

This paper is concerned with the problem of stabilizing continuous-time linear time-invariant systems subject to quantization and Denial-of-Service (DoS) attacks. In this context, two DoS-induced challenges emerge with the design of resilient encoding schemes, namely, the coupling between encoding strategies of different signals, and the synchronization between the encoder and decoder. To address these challenges, a novel structure that is equipped with a deadbeat controller as well as a delicate transmission protocol for the input and output channels, co-designed leveraging the controllability index, is put forward. When both input and output channels are subject to DoS attacks and quantization, the proposed structure is shown able to decouple the encoding schemes for input, output, and estimated output signals. This property is further corroborated by designing encoding schemes as well as conditions that ensure exponential stability of the closed-loop system. On the other hand, when only the output channel is subject to network phenomenon, the proposed structure can achieve exponential stabilization without acknowledgment (ACK) signals, in contrast to existing ACK-based results. Finally, a numerical example is given to demonstrate the practical merits of the proposed approach as well as the theory.

eess.SY↗

arXiv2021

Training Networks in Null Space of Feature Covariance for Continual Learning

In the setting of continual learning, a network is trained on a sequence of tasks, and suffers from catastrophic forgetting. To balance plasticity and stability of network in continual learning, in this paper, we propose a novel network training algorithm called Adam-NSCL, which sequentially optimizes network parameters in the null space of previous tasks. We first propose two mathematical conditions respectively for achieving network stability and plasticity in continual learning. Based on them, the network training for sequential tasks can be simply achieved by projecting the candidate parameter update into the approximate null space of all previous tasks in the network training process, where the candidate parameter update can be generated by Adam. The approximate null space can be derived by applying singular value decomposition to the uncentered covariance matrix of all input features of previous tasks for each linear layer. For efficiency, the uncentered covariance matrix can be incrementally computed after learning each task. We also empirically verify the rationality of the approximate null space at each linear layer. We apply our approach to training networks for continual learning on benchmark datasets of CIFAR-100 and TinyImageNet, and the results suggest that the proposed approach outperforms or matches the state-ot-the-art continual learning approaches.

cs.LG↗

arXiv2021

You Only Look One-level Feature

This paper revisits feature pyramids networks (FPN) for one-stage detectors and points out that the success of FPN is due to its divide-and-conquer solution to the optimization problem in object detection rather than multi-scale feature fusion. From the perspective of optimization, we introduce an alternative way to address the problem instead of adopting the complex feature pyramids - {\em utilizing only one-level feature for detection}. Based on the simple and efficient solution, we present You Only Look One-level Feature (YOLOF). In our method, two key components, Dilated Encoder and Uniform Matching, are proposed and bring considerable improvements. Extensive experiments on the COCO benchmark prove the effectiveness of the proposed model. Our YOLOF achieves comparable results with its feature pyramids counterpart RetinaNet while being $2.5\times$ faster. Without transformer layers, YOLOF can match the performance of DETR in a single-level feature manner with $7\times$ less training epochs. With an image size of $608\times608$, YOLOF achieves 44.3 mAP running at 60 fps on 2080Ti, which is $13\%$ faster than YOLOv4. Code is available at \url{https://github.com/megvii-model/YOLOF}.

cs.CV↗

arXiv2021

Dynamic Region-Aware Convolution

We propose a new convolution called Dynamic Region-Aware Convolution (DRConv), which can automatically assign multiple filters to corresponding spatial regions where features have similar representation. In this way, DRConv outperforms standard convolution in modeling semantic variations. Standard convolutional layer can increase the number of filers to extract more visual elements but results in high computational cost. More gracefully, our DRConv transfers the increasing channel-wise filters to spatial dimension with learnable instructor, which not only improve representation ability of convolution, but also maintains computational cost and the translation-invariance as standard convolution dose. DRConv is an effective and elegant method for handling complex and variable spatial information distribution. It can substitute standard convolution in any existing networks for its plug-and-play property, especially to power convolution layers in efficient networks. We evaluate DRConv on a wide range of models (MobileNet series, ShuffleNetV2, etc.) and tasks (Classification, Face Recognition, Detection and Segmentation). On ImageNet classification, DRConv-based ShuffleNetV2-0.5x achieves state-of-the-art performance of 67.1% at 46M multiply-adds level with 6.3% relative improvement.

cs.CV↗

arXiv2021

Dynamic Scale Training for Object Detection

We propose a Dynamic Scale Training paradigm (abbreviated as DST) to mitigate scale variation challenge in object detection. Previous strategies like image pyramid, multi-scale training, and their variants are aiming at preparing scale-invariant data for model optimization. However, the preparation procedure is unaware of the following optimization process that restricts their capability in handling the scale variation. Instead, in our paradigm, we use feedback information from the optimization process to dynamically guide the data preparation. The proposed method is surprisingly simple yet obtains significant gains (2%+ Average Precision on MS COCO dataset), outperforming previous methods. Experimental results demonstrate the efficacy of our proposed DST method towards scale variation handling. It could also generalize to various backbones, benchmarks, and other challenging downstream tasks like instance segmentation. It does not introduce inference overhead and could serve as a free lunch for general detection configurations. Besides, it also facilitates efficient training due to fast convergence. Code and models are available at github.com/yukang2017/Stitcher.

cs.CV↗

arXiv2021

End-to-End Human Object Interaction Detection with HOI Transformer

We propose HOI Transformer to tackle human object interaction (HOI) detection in an end-to-end manner. Current approaches either decouple HOI task into separated stages of object detection and interaction classification or introduce surrogate interaction problem. In contrast, our method, named HOI Transformer, streamlines the HOI pipeline by eliminating the need for many hand-designed components. HOI Transformer reasons about the relations of objects and humans from global image context and directly predicts HOI instances in parallel. A quintuple matching loss is introduced to force HOI predictions in a unified way. Our method is conceptually much simpler and demonstrates improved accuracy. Without bells and whistles, HOI Transformer achieves $26.61\% $ $ AP $ on HICO-DET and $52.9\%$ $AP_{role}$ on V-COCO, surpassing previous methods with the advantage of being much simpler. We hope our approach will serve as a simple and effective alternative for HOI tasks. Code is available at https://github.com/bbepoch/HoiTransformer .

cs.CV↗

arXiv2021

Topological insulators in the NaCaBi family with large SOC gaps

By means of first-principles calculations and crystal structure searching techniques, we predict that a new NaCaBi family crystallized into the ZrBeSi-type structure (\ie $P6_{3}/mmc$) are strong topological insulators (STIs). Taking $P6_{3}/mmc$ NaCaBi as an example, the calculated band structure indicates that there is a band inversion between two opposite-parity bands at the $Γ$ point. In contrast to the well-known Bi$_2$Se$_3$ family, the band inversion in the NaCaBi family has already occured even without spin-orbit coupling (SOC), giving rise to a nodal ring surrounding $Γ$ in the $k_z=0$ plane (protected by $M_z$ symmetry). With time reversal symmetry $(\cal T)$ and inversion symmetry $(\cal I)$, the spinless nodal-line metallic phase protected by $[{\cal TI}]^2=1$ is the weak-SOC limit of the spinful topological insulating phase. Upon including SOC, the nodal ring is gapped, driving the system into a STI. Besides inversion symmetry, the nontrivial topology of NaCaBi can also be indicated by $\bar{6}$ symmetry. More surprisingly, the SOC-induced band gap in NaCaBi is about 0.34 eV, which is larger than the energy scale of room temperature. Four other compounds (KBaBi, KSrBi, RbBaBi and RbSrBi) in the family are stable at ambient pressure, both in thermodynamics and lattice dynamics, even though the gaps of them are smaller than that of NaCaBi. Thus, they provide good platforms to study topological states both in theory and experiments.

cond-mat.mtrl-sci↗

arXiv2021

FFB6D: A Full Flow Bidirectional Fusion Network for 6D Pose Estimation

In this work, we present FFB6D, a Full Flow Bidirectional fusion network designed for 6D pose estimation from a single RGBD image. Our key insight is that appearance information in the RGB image and geometry information from the depth image are two complementary data sources, and it still remains unknown how to fully leverage them. Towards this end, we propose FFB6D, which learns to combine appearance and geometry information for representation learning as well as output representation selection. Specifically, at the representation learning stage, we build bidirectional fusion modules in the full flow of the two networks, where fusion is applied to each encoding and decoding layer. In this way, the two networks can leverage local and global complementary information from the other one to obtain better representations. Moreover, at the output representation stage, we designed a simple but effective 3D keypoints selection algorithm considering the texture and geometry information of objects, which simplifies keypoint localization for precise pose estimation. Experimental results show that our method outperforms the state-of-the-art by large margins on several benchmarks. Code and video are available at \url{https://github.com/ethnhe/FFB6D.git}.

cs.CV↗

arXiv2021

Distributed proximal gradient algorithm for non-smooth non-convex optimization over time-varying networks

This note studies the distributed non-convex optimization problem with non-smooth regularization, which has wide applications in decentralized learning, estimation and control. The objective function is the sum of different local objective functions, which consist of differentiable (possibly non-convex) cost functions and non-smooth convex functions. This paper presents a distributed proximal gradient algorithm for the non-smooth non-convex optimization problem over time-varying multi-agent networks. Each agent updates local variable estimate by the multi-step consensus operator and the proximal operator. We prove that the generated local variables achieve consensus and converge to the set of critical points with convergence rate $O(1/T)$. Finally, we verify the efficacy of proposed algorithm by numerical simulations.

math.OC↗

arXiv2021

Enhancing Crystal Structure Prediction by decomposition methods based on graph theory

Crystal structure prediction algorithms have become powerful tools for materials discovery in recent years, however, they are usually limited to relatively small systems. The main challenge is that the number of local minima grows exponentially with system size. In this work, we proposed two crossover-mutation schemes based on graph theory to accelerate the evolutionary structure searching. These schemes can detect molecules or clusters inside periodic networks using quotient graphs for crystals and the decomposition can dramatically reduce the searching space. Sufficient examples for the test, including the high pressure phases of methane, ammonia, MgAl2O4, and boron, show that these new evolution schemes can obviously improve the success rate and searching efficiency compared with the standard method in both isolated and extended systems.

cond-mat.mtrl-sci↗

arXiv2021

Negative linear compressibility and unusual dynamic behaviors of NaB3

First-principles calculations reveal that sodium boride (NaB3) undergoes a phase transition from a tetragonal P4/mbm phase to an orthorhombic Pbam phase at about 16 GPa, accompanied by counterintuitive lattice expansion along the crystallographic a-axis. This unusual compression behavior is identified as negative linear compressibility (NLC), which is dominantly attributed to the symmetry-breaking of boron framework. Meanwhile, the P4/mbm and Pbam phases form superionic conductors after undergoing a peculiar swap state at high temperature. Specifically, under warm conditions the Na cation pairs exhibit a rare local exchange (or rotation) behavior, which may be originated from the asymmetric energy barriers of different diffusion paths. The study of NaB3 compound sheds new light on a material with the combination of NLC and ion transportation at extreme conditions.

cond-mat.mtrl-sci↗

arXiv2021

Using Long Short-Term Memory (LSTM) and Internet of Things (IoT) for localized surface temperature forecasting in an urban environment

The rising temperature is one of the key indicators of a warming climate, and it can cause extensive stress to biological systems as well as built structures. Due to the heat island effect, it is most severe in urban environments compared to other landscapes due to the decrease in vegetation associated with a dense human-built environment. It is essential to adequately monitor the local temperature dynamics to mitigate risks associated with increasing temperatures, which can include short term strategy to protect people and animals, to long term strategy to how to build a new structure and cope with extreme events. Observed temperature is also a very important input for atmospheric models, and accurate data can lead to better future forecasts. Ambient temperature collected at ground level can have a higher variability when compared to regional weather forecasts, which fail to capture the local dynamics. There remains a clear need for an accurate air temperature prediction at the sub-urban scale at high temporal and spatial resolution. This research proposed a framework based on Long Short-Term Memory (LSTM) deep learning network to generate day-ahead hourly temperature forecast with high spatial resolution. A case study is shown which uses historical in-situ observations and Internet of Things (IoT) observations for New York City, USA. By leveraging the historical air temperature data from in-situ observations, the LSTM model can be exposed to more historical patterns that might not be present in the IoT observations. Meanwhile, by using IoT observations, the spatial resolution of air temperature predictions is significantly improved.

cs.LG↗

arXiv2021

Numerical analysis of a deep learning formulation of elastic full waveform inversion with high order total variation regularization in different parameterization

We have formulated elastic seismic full waveform inversion (FWI) within a deep learning environment. In our formulation, a recurrent neural network is set up with rules enforcing elastic wave propagation, with the wavefield projected onto a measurement surface acting as the synthetic data to be compared with observed seismic data. Gradients for iterative updating of an elastic model, with a variety of parameterizations and misfit functionals, can be efficiently constructed within the network through the automatic differential method. With this method, the inversion based on complex misfits can be calculated. We survey the impact of different complex misfits (based on the l2, l1 ) with high order total variation (TV) regulations on multiparameter elastic FWI recovery of models within velocity/density, modulus/density, and stiffness parameter/density parameterizations. We analyze parameter cross-talk. Inversion results on simple and complex models show that the RNN elastic FWI with high order TV regulation using l1 norm can help mitigate cross-talk issues with gradient-based optimization methods.

physics.geo-ph↗

arXiv2021

Momentum^2 Teacher: Momentum Teacher with Momentum Statistics for Self-Supervised Learning

In this paper, we present a novel approach, Momentum$^2$ Teacher, for student-teacher based self-supervised learning. The approach performs momentum update on both network weights and batch normalization (BN) statistics. The teacher's weight is a momentum update of the student, and the teacher's BN statistics is a momentum update of those in history. The Momentum$^2$ Teacher is simple and efficient. It can achieve the state of the art results (74.5\%) under ImageNet linear evaluation protocol using small-batch size(\eg, 128), without requiring large-batch training on special hardware like TPU or inefficient across GPU operation (\eg, shuffling BN, synced BN). Our implementation and pre-trained models will be given on GitHub\footnote{https://github.com/zengarden/momentum2-teacher}.

cs.LG↗

arXiv2021

Van Hove Singularity Arising from Mexican-Hat-Shaped Inverted Bands in the Topological Insulator Sn-doped Bi$_{1.1}$Sb$_{0.9}$Te$_{2}$S

The optical properties of Sn-doped Bi$_{1.1}$Sb$_{0.9}$Te$_{2}$S, the most bulk-insulating topological insulator thus far, have been examined at different temperatures over a broad frequency range. No Drude response is detected in the low-frequency range down to 30~cm$^{-1}$, corroborating the excellent bulk-insulating property of this material. Intriguingly, we observe a sharp peak at about 2\,200~cm$^{-1}$ in the optical conductivity at 5~K. Further quantitative analyses of the line shape and temperature dependence of this sharp peak, in combination with first-principles calculations, suggest that it corresponds to a van Hove singularity arising from Mexican-hat-shaped inverted bands. Such a van Hove singularity is a pivotal ingredient of various strongly correlated phases.

cond-mat.str-el↗

arXiv2021

Dynamic Hybrid Relation Network for Cross-Domain Context-Dependent Semantic Parsing

Semantic parsing has long been a fundamental problem in natural language processing. Recently, cross-domain context-dependent semantic parsing has become a new focus of research. Central to the problem is the challenge of leveraging contextual information of both natural language utterance and database schemas in the interaction history. In this paper, we present a dynamic graph framework that is capable of effectively modelling contextual utterances, tokens, database schemas, and their complicated interaction as the conversation proceeds. The framework employs a dynamic memory decay mechanism that incorporates inductive bias to integrate enriched contextual relation representation, which is further enhanced with a powerful reranking model. At the time of writing, we demonstrate that the proposed framework outperforms all existing models by large margins, achieving new state-of-the-art performance on two large-scale benchmarks, the SParC and CoSQL datasets. Specifically, the model attains a 55.8% question-match and 30.8% interaction-match accuracy on SParC, and a 46.8% question-match and 17.0% interaction-match accuracy on CoSQL.

cs.CL↗

arXiv2020

Implicit Feature Pyramid Network for Object Detection

In this paper, we present an implicit feature pyramid network (i-FPN) for object detection. Existing FPNs stack several cross-scale blocks to obtain large receptive field. We propose to use an implicit function, recently introduced in deep equilibrium model (DEQ), to model the transformation of FPN. We develop a residual-like iteration to updates the hidden states efficiently. Experimental results on MS COCO dataset show that i-FPN can significantly boost detection performance compared to baseline detectors with ResNet-50-FPN: +3.4, +3.2, +3.5, +4.2, +3.2 mAP on RetinaNet, Faster-RCNN, FCOS, ATSS and AutoAssign, respectively.

cs.CV↗

arXiv2020

Efficient Human Pose Estimation by Learning Deeply Aggregated Representations

In this paper, we propose an efficient human pose estimation network (DANet) by learning deeply aggregated representations. Most existing models explore multi-scale information mainly from features with different spatial sizes. Powerful multi-scale representations usually rely on the cascaded pyramid framework. This framework largely boosts the performance but in the meanwhile makes networks very deep and complex. Instead, we focus on exploiting multi-scale information from layers with different receptive-field sizes and then making full of use this information by improving the fusion method. Specifically, we propose an orthogonal attention block (OAB) and a second-order fusion unit (SFU). The OAB learns multi-scale information from different layers and enhances them by encouraging them to be diverse. The SFU adaptively selects and fuses diverse multi-scale information and suppress the redundant ones. This could maximize the effective information in final fused representations. With the help of OAB and SFU, our single pyramid network may be able to generate deeply aggregated representations that contain even richer multi-scale information and have a larger representing capacity than that of cascaded networks. Thus, our networks could achieve comparable or even better accuracy with much smaller model complexity. Specifically, our \mbox{DANet-72} achieves $70.5$ in AP score on COCO test-dev set with only $1.0G$ FLOPs. Its speed on a CPU platform achieves $58$ Persons-Per-Second~(PPS).

cs.CV↗