arXiv subjects

Dong Li

Publications and source records attributed to Dong Li.

At least 451 records · Page 25Linked to original sources

arXiv2021

Scattering-lens based quantum imaging beyond shot noise

The scheme of optical imaging using scattering lens can provide a resolution beyond the classical optical diffraction limit with a coherent-state input. Nevertheless, due to the shot noise of the coherent state, the corresponding signal-to-noise ratio and resolution are both still shot-noise-limited. In order to circumvent this problem, we theoretically propose an alternative scheme where the squeezed state (with a sub-shot noise) is considered as input and the quantum noise is then suppressed below the shot-noise level. Consequently, when comparing with the previous imaging scheme (using combination of coherent state and scattering lens), our proposal is able to achieve an enhanced signal-to-noise ratio for a given scattering lens. Meanwhile, it is demonstrated that the resolution is also improved. We believe that this method may afford a new way of using squeezed states and enable a higher performance than that of using coherent state and scattering lens.

quant-ph↗

arXiv2021

Probabilistic Multi-View Fusion of Active Stereo Depth Maps for Robotic Bin-Picking

The reliable fusion of depth maps from multiple viewpoints has become an important problem in many 3D reconstruction pipelines. In this work, we investigate its impact on robotic bin-picking tasks such as 6D object pose estimation. The performance of object pose estimation relies heavily on the quality of depth data. However, due to the prevalence of shiny surfaces and cluttered scenes, industrial grade depth cameras often fail to sense depth or generate unreliable measurements from a single viewpoint. To this end, we propose a novel probabilistic framework for scene reconstruction in robotic bin-picking. Based on active stereo camera data, we first explicitly estimate the uncertainty of depth measurements for mitigating the adverse effects of both noise and outliers. The uncertainty estimates are then incorporated into a probabilistic model for incrementally updating the scene. To extensively evaluate the traditional fusion approach alongside our own approach, we will release a novel representative dataset with multiple views for each bin and curated parts. Over the entire dataset, we demonstrate that our framework outperforms a traditional fusion approach by a 12.8% reduction in reconstruction error, and 6.1% improvement in detection rate. The dataset will be available at https://www.trailab.utias.utoronto.ca/robi.

cs.RO↗

arXiv2021

Learning Symbolic Rules for Interpretable Deep Reinforcement Learning

Recent progress in deep reinforcement learning (DRL) can be largely attributed to the use of neural networks. However, this black-box approach fails to explain the learned policy in a human understandable way. To address this challenge and improve the transparency, we propose a Neural Symbolic Reinforcement Learning framework by introducing symbolic logic into DRL. This framework features a fertilization of reasoning and learning modules, enabling end-to-end learning with prior symbolic knowledge. Moreover, interpretability is achieved by extracting the logical rules learned by the reasoning module in a symbolic rule space. The experimental results show that our framework has better interpretability, along with competing performance in comparison to state-of-the-art approaches.

cs.AI↗

arXiv2021

TRUST: Triangle Counting Reloaded on GPUs

Triangle counting is a building block for a wide range of graph applications. Traditional wisdom suggests that i) hashing is not suitable for triangle counting, ii) edge-centric triangle counting beats vertex-centric design, and iii) communication-free and workload balanced graph partitioning is a grand challenge for triangle counting. On the contrary, we advocate that i) hashing can help the key operations for scalable triangle counting on Graphics Processing Units (GPUs), i.e., list intersection and graph partitioning, ii)vertex-centric design reduces both hash table construction cost and memory consumption, which is limited on GPUs. In addition, iii) we exploit graph and workload collaborative, and hashing-based 2D partitioning to scale vertex-centric triangle counting over 1,000 GPUswith sustained scalability. In this work, we present TRUST which performs triangle counting with the hash operation and vertex-centric mechanism at the core. To the best of our knowledge, TRUSTis the first work that achieves over one trillion Traversed Edges Per Second (TEPS) rate for triangle counting.

cs.DC↗

arXiv2021

Strong but Simple Baseline with Dual-Granularity Triplet Loss for Visible-Thermal Person Re-Identification

In this letter, we propose a conceptually simple and effective dual-granularity triplet loss for visible-thermal person re-identification (VT-ReID). In general, ReID models are always trained with the sample-based triplet loss and identification loss from the fine granularity level. It is possible when a center-based loss is introduced to encourage the intra-class compactness and inter-class discrimination from the coarse granularity level. Our proposed dual-granularity triplet loss well organizes the sample-based triplet loss and center-based triplet loss in a hierarchical fine to coarse granularity manner, just with some simple configurations of typical operations, such as pooling and batch normalization. Experiments on RegDB and SYSU-MM01 datasets show that with only the global features our dual-granularity triplet loss can improve the VT-ReID performance by a significant margin. It can be a strong VT-ReID baseline to boost future research with high quality.

cs.CV↗

arXiv2021

On Estimating Recommendation Evaluation Metrics under Sampling

Since the recent study (Krichene and Rendle 2020) done by Krichene and Rendle on the sampling-based top-k evaluation metric for recommendation, there has been a lot of debates on the validity of using sampling to evaluate recommendation algorithms. Though their work and the recent work (Li et al.2020) have proposed some basic approaches for mapping the sampling-based metrics to their global counterparts which rank the entire set of items, there is still a lack of understanding and consensus on how sampling should be used for recommendation evaluation. The proposed approaches either are rather uninformative (linking sampling to metric evaluation) or can only work on simple metrics, such as Recall/Precision (Krichene and Rendle 2020; Li et al. 2020). In this paper, we introduce a new research problem on learning the empirical rank distribution, and a new approach based on the estimated rank distribution, to estimate the top-k metrics. Since this question is closely related to the underlying mechanism of sampling for recommendation, tackling it can help better understand the power of sampling and can help resolve the questions of if and how should we use sampling for evaluating recommendation. We introduce two approaches based on MLE (MaximalLikelihood Estimation) and its weighted variants, and ME(Maximal Entropy) principals to recover the empirical rank distribution, and then utilize them for metrics estimation. The experimental results show the advantages of using the new approaches for evaluating recommendation algorithms based on top-k metrics.

cs.IR↗

arXiv2021

MATCH: An MPI Fault Tolerance Benchmark Suite

MPI has been ubiquitously deployed in flagship HPC systems aiming to accelerate distributed scientific applications running on tens of hundreds of processes and compute nodes. Maintaining the correctness and integrity of MPI application execution is critical, especially for safety-critical scientific applications. Therefore, a collection of effective MPI fault tolerance techniques have been proposed to enable MPI application execution to efficiently resume from system failures. However, there is no structured way to study and compare different MPI fault tolerance designs, so to guide the selection and development of efficient MPI fault tolerance techniques for distinct scenarios. To solve this problem, we design, develop, and evaluate a benchmark suite called MATCH to characterize, research, and comprehensively compare different combinations and configurations of MPI fault tolerance designs. Our investigation derives useful findings: (1) Reinit recovery in general performs better than ULFM recovery; (2) Reinit recovery is independent of the scaling size and the input problem size, whereas ULFM recovery is not; (3) Using Reinit recovery with FTI checkpointing is a highly efficient fault tolerance design. MATCH code is available at https://github.com/kakulo/MPI- FT- Bench.

cs.DC↗

arXiv2021

MOARD: Modeling Application Resilience to Transient Faults on Data Objects

Understanding application resilience (or error tolerance) in the presence of hardware transient faults on data objects is critical to ensure computing integrity and enable efficient application-level fault tolerance mechanisms. However, we lack a method and a tool to quantify application resilience to transient faults on data objects. The traditional method, random fault injection, cannot help, because of losing data semantics and insufficient information on how and where errors are tolerated. In this paper, we introduce a method and a tool (called MOARD) to model and quantify application resilience to transient faults on data objects. Our method is based on systematically quantifying error masking events caused by application-inherent semantics and program constructs. We use MOARD to study how and why errors in data objects can be tolerated by the application. We demonstrate tangible benefits of using MOARD to direct a fault tolerance mechanism to protect data objects.

cs.DC↗

arXiv2021

Diagnosing a Solar Flaring Core with Bidirectional Quasi-Periodic Fast Propagating Magnetoacoustic Waves

Quasi-periodic fast propagating (QFP) waves are often excited by solar flares, and could be trapped in the coronal structure with low Alfvén speed, so they could be used as a diagnosing tool for both the flaring core and magnetic waveguide. As the periodicity of a QFP wave could originate from a periodic source or be dispersively waveguided, it is a key parameter for diagnosing the flaring core and waveguide. In this paper, we study two QFP waves excited by a GOES-class C1.3 solar flare occurring at active region NOAA 12734 on 8 March 2019. Two QFP waves were guided by two oppositely oriented coronal funnel. The periods of two QFP waves were identical and were roughly equal to the period of the oscillatory signal in the X-ray and 17 GHz radio emission released by the flaring core. It is very likely that the two QFP waves could be periodically excited by the flaring core. Many features of this QFP wave event is consistent with the magnetic tuning fork model. We also investigated the seismological application with QFP waves, and found that the magnetic field inferred with magnetohydrodynamic seismology was consistent with that obtained in magnetic extrapolation model. Our study suggest that the QFP wave is a good tool for diagnosing both the flaring core and the magnetic waveguide.

astro-ph.SR↗

arXiv2021

Excess-noise suppression for a squeezed state propagating through random amplifying media via wave-front shaping

After propagating through a random amplifying medium, a squeezed state commonly shows excess noise above the shot-noise level. Since large noise can significantly reduce the signal-to-noise ratio, it is detrimental for precision measurement. To circumvent this problem, we propose a noise-reduction scheme using wavefront shaping. It is demonstrated that the average output quantum noise can be effectively suppressed even beyond the shot-noise limit. Both the decrease on amplification strength and the increase on input squeezing strength can give rise to a decrease in the suppressed average quantum noise. Our results not only show the feasibility of manipulating the output quantum noise of random amplifying media, but also indicate potential applications in quantum information processing in complex environments, such as, quantum imaging, quantum communication, and quantum key distribution.

quant-ph↗

arXiv2021

ZeRO-Offload: Democratizing Billion-Scale Model Training

Large-scale model training has been a playing ground for a limited few requiring complex model refactoring and access to prohibitively expensive GPU clusters. ZeRO-Offload changes the large model training landscape by making large model training accessible to nearly everyone. It can train models with over 13 billion parameters on a single GPU, a 10x increase in size compared to popular framework such as PyTorch, and it does so without requiring any model change from the data scientists or sacrificing computational efficiency. ZeRO-Offload enables large model training by offloading data and compute to CPU. To preserve compute efficiency, it is designed to minimize the data movement to/from GPU, and reduce CPU compute time while maximizing memory savings on GPU. As a result, ZeRO-Offload can achieve 40 TFlops/GPU on a single NVIDIA V100 GPU for 10B parameter model compared to 30TF using PyTorch alone for a 1.4B parameter model, the largest that can be trained without running out of memory. ZeRO-Offload is also designed to scale on multiple-GPUs when available, offering near linear speedup on up to 128 GPUs. Additionally, it can work together with model parallelism to train models with over 70 billion parameters on a single DGX-2 box, a 4.5x increase in model size compared to using model parallelism alone. By combining compute and memory efficiency with ease-of-use, ZeRO-Offload democratizes large-scale model training making it accessible to even data scientists with access to just a single GPU.

cs.DC↗

arXiv2021

Solving phase retrieval with random initial guess is nearly as good as by spectral initialization

The problem of recovering a signal $\mathbf{x}\in \mathbb{R}^n$ from a set of magnitude measurements $y_i=|\langle \mathbf{a}_i, \mathbf{x} \rangle |, \; i=1,\ldots,m$ is referred as phase retrieval, which has many applications in fields of physical sciences and engineering. In this paper we show that the smoothed amplitude flow model for phase retrieval has benign geometric structure under the optimal sampling complexity. In particular, we show that when the measurements $\mathbf{a}_i\in \mathbb{R}^n$ are Gaussian random vectors and the number of measurements $m\ge Cn$, our smoothed amplitude flow model has no spurious local minimizers with high probability, ie., the target solution $\mathbf{x}$ is the unique global minimizer (up to a global phase) and the loss function has a negative directional curvature around each saddle point. Due to this benign geometric landscape, the phase retrieval problem can be solved by the gradient descent algorithms without spectral initialization. Numerical experiments show that the gradient descent algorithm with random initialization performs well even comparing with state-of-the-art algorithms with spectral initialization in empirical success rate and convergence speed.

cs.IT↗

arXiv2020

Towards Effective Context for Meta-Reinforcement Learning: an Approach based on Contrastive Learning

Context, the embedding of previous collected trajectories, is a powerful construct for Meta-Reinforcement Learning (Meta-RL) algorithms. By conditioning on an effective context, Meta-RL policies can easily generalize to new tasks within a few adaptation steps. We argue that improving the quality of context involves answering two questions: 1. How to train a compact and sufficient encoder that can embed the task-specific information contained in prior trajectories? 2. How to collect informative trajectories of which the corresponding context reflects the specification of tasks? To this end, we propose a novel Meta-RL framework called CCM (Contrastive learning augmented Context-based Meta-RL). We first focus on the contrastive nature behind different tasks and leverage it to train a compact and sufficient context encoder. Further, we train a separate exploration policy and theoretically derive a new information-gain-based objective which aims to collect informative trajectories in a few steps. Empirically, we evaluate our approaches on common benchmarks as well as several complex sparse-reward environments. The experimental results show that CCM outperforms state-of-the-art algorithms by addressing previously mentioned problems respectively.

cs.LG↗

arXiv2020

Towards Better Object Detection in Scale Variation with Adaptive Feature Selection

It is a common practice to exploit pyramidal feature representation to tackle the problem of scale variation in object instances. However, most of them still predict the objects in a certain range of scales based solely or mainly on a single-level representation, yielding inferior detection performance. To this end, we propose a novel adaptive feature selection module (AFSM), to automatically learn the way to fuse multi-level representations in the channel dimension, in a data-driven manner. It significantly improves the performance of the detectors that have a feature pyramid structure, while introducing nearly free inference overhead. Moreover, a class-aware sampling mechanism (CASM) is proposed to tackle the class imbalance problem, by re-weighting the sampling ratio to each of the training images, based on the statistical characteristics of each class. This is crucial to improve the performance of the minor classes. Experimental results demonstrate the effectiveness of the proposed method, with 83.04% mAP at 15.96 FPS on the VOC dataset, and 39.48% AP on the VisDrone-DET validation subset, respectively, outperforming other state-of-the-art detectors considerably. The code is available at https://github.com/ZeHuiGong/AFSM.git.

cs.CV↗

arXiv2020

Raman Study of Cooper Pairing Instabilities in (Li$_{1-x}$Fe$_x$)OHFeSe

We studied the electronic Raman spectra of (Li$_{1-x}$Fe$_x$)OHFeSe as a function of light polarization and temperature. In the B$_{1g}$ spectra alone we observe the redistribution of spectral weight expected for a superconductor and two well-resolved peaks below T$_c$. The nearly resolution-limited peak at 110 cm$^{-1}$ (13.6 meV) is identified as a collective mode. The peak at 190 cm$^{-1}$ (23.6 meV) is presumably another collective mode since the line is symmetric and its energy is significantly below the gap energy observed by single-particle spectroscopies. Given the experimental band structure of (Li$_{1-x}$Fe$_x$)OHFeSe, the most plausible explanations include conventional spin-fluctuation pairing between the electron bands and the incipient hole band and pairing between the hybridized electron bands. The absence of gap features in A$_{1g}$ and B$_{2g}$ symmetry favors the second case. Thus, in spite of various differences between the pnictides and chalcogenides, this Letter demonstrates the proximity of pairing states and the importance of band structure effects in the Fe-based compounds.

cond-mat.supr-con↗

arXiv2020

Towards Optimal Filter Pruning with Balanced Performance and Pruning Speed

Filter pruning has drawn more attention since resource constrained platform requires more compact model for deployment. However, current pruning methods suffer either from the inferior performance of one-shot methods, or the expensive time cost of iterative training methods. In this paper, we propose a balanced filter pruning method for both performance and pruning speed. Based on the filter importance criteria, our method is able to prune a layer with approximate layer-wise optimal pruning rate at preset loss variation. The network is pruned in the layer-wise way without the time consuming prune-retrain iteration. If a pre-defined pruning rate for the entire network is given, we also introduce a method to find the corresponding loss variation threshold with fast converging speed. Moreover, we propose the layer group pruning and channel selection mechanism for channel alignment in network with short connections. The proposed pruning method is widely applicable to common architectures and does not involve any additional training except the final fine-tuning. Comprehensive experiments show that our method outperforms many state-of-the-art approaches.

cs.CV↗

arXiv2020

Adaptive inference for a semiparametric generalized autoregressive conditional heteroskedasticity model

This paper considers a semiparametric generalized autoregressive conditional heteroskedasticity (S-GARCH) model. For this model, we first estimate the time-varying long run component for unconditional variance by the kernel estimator, and then estimate the non-time-varying parameters in GARCH-type short run component by the quasi maximum likelihood estimator (QMLE). We show that the QMLE is asymptotically normal with the parametric convergence rate. Next, we construct a Lagrange multiplier test for linear parameter constraint and a portmanteau test for model checking, and obtain their asymptotic null distributions. Our entire statistical inference procedure works for the non-stationary data with two important features: first, our QMLE and two tests are adaptive to the unknown form of the long run component; second, our QMLE and two tests share the same efficiency and testing power as those in variance targeting method when the S-GARCH model is stationary.

stat.ME↗

arXiv2020

Electronic phase diagram of iron chalcogenide superconductors FeSe1-xSx and FeSe1-yTey

Here we establish a combined electronic phase diagram of isoelectronic FeSe1-xSx (0.19 > x > 0.0) and FeSe1-yTey (0.04 < y < 1.0) single crystals. The FeSe1-yTey crystals with y = 0.04 - 0.30 are grown by a hydrothermal ion-deintercalation (HID) method. Based on combined experiments of the specific heat, electrical transport, and angle-resolved photoemission spectroscopy, no signature of the tetragonal-symmetry-broken transition to orthorhombic (nematic) phase is observed in the HID FeSe1-yTey samples, as compared with the FeSe1-xSx samples showing this transition at Ts. A ubiquitous dip-like temperature dependence of the Hall coefficient is observed around a characteristic temperature T* in the tetragonal regimes, which is well above the superconducting transition. More importantly, we find that the superconducting transition temperature Tc is positively correlated with the Hall-dip temperature T* across the FeSe1-xSx and FeSe1-yTey systems, suggesting that the tetragonal background is a fundamental host for the superconductivity.

cond-mat.supr-con↗