arXiv subjects

Shuai Li

Publications and source records attributed to Shuai Li.

At least 271 records · Page 15Linked to original sources

arXiv2023

One-to-Few Label Assignment for End-to-End Dense Detection

One-to-one (o2o) label assignment plays a key role for transformer based end-to-end detection, and it has been recently introduced in fully convolutional detectors for end-to-end dense detection. However, o2o can degrade the feature learning efficiency due to the limited number of positive samples. Though extra positive samples are introduced to mitigate this issue in recent DETRs, the computation of self- and cross- attentions in the decoder limits its practical application to dense and fully convolutional detectors. In this work, we propose a simple yet effective one-to-few (o2f) label assignment strategy for end-to-end dense detection. Apart from defining one positive and many negative anchors for each object, we define several soft anchors, which serve as positive and negative samples simultaneously. The positive and negative weights of these soft anchors are dynamically adjusted during training so that they can contribute more to ``representation learning'' in the early training stage, and contribute more to ``duplicated prediction removal'' in the later stage. The detector trained in this way can not only learn a strong feature representation but also perform end-to-end dense detection. Experiments on COCO and CrowdHuman datasets demonstrate the effectiveness of the o2f scheme. Code is available at https://github.com/strongwolf/o2f.

cs.CV↗

arXiv2023

MSF: Motion-guided Sequential Fusion for Efficient 3D Object Detection from Point Cloud Sequences

Point cloud sequences are commonly used to accurately detect 3D objects in applications such as autonomous driving. Current top-performing multi-frame detectors mostly follow a Detect-and-Fuse framework, which extracts features from each frame of the sequence and fuses them to detect the objects in the current frame. However, this inevitably leads to redundant computation since adjacent frames are highly correlated. In this paper, we propose an efficient Motion-guided Sequential Fusion (MSF) method, which exploits the continuity of object motion to mine useful sequential contexts for object detection in the current frame. We first generate 3D proposals on the current frame and propagate them to preceding frames based on the estimated velocities. The points-of-interest are then pooled from the sequence and encoded as proposal features. A novel Bidirectional Feature Aggregation (BiFA) module is further proposed to facilitate the interactions of proposal features across frames. Besides, we optimize the point cloud pooling by a voxel-based sampling technique so that millions of points can be processed in several milliseconds. The proposed MSF method achieves not only better efficiency than other multi-frame detectors but also leading accuracy, with 83.12% and 78.30% mAP on the LEVEL1 and LEVEL2 test sets of Waymo Open Dataset, respectively. Codes can be found at \url{https://github.com/skyhehe123/MSF}.

cs.CV↗

arXiv2023

DynaMask: Dynamic Mask Selection for Instance Segmentation

The representative instance segmentation methods mostly segment different object instances with a mask of the fixed resolution, e.g., 28*28 grid. However, a low-resolution mask loses rich details, while a high-resolution mask incurs quadratic computation overhead. It is a challenging task to predict the optimal binary mask for each instance. In this paper, we propose to dynamically select suitable masks for different object proposals. First, a dual-level Feature Pyramid Network (FPN) with adaptive feature aggregation is developed to gradually increase the mask grid resolution, ensuring high-quality segmentation of objects. Specifically, an efficient region-level top-down path (r-FPN) is introduced to incorporate complementary contextual and detailed information from different stages of image-level FPN (i-FPN). Then, to alleviate the increase of computation and memory costs caused by using large masks, we develop a Mask Switch Module (MSM) with negligible computational cost to select the most suitable mask resolution for each instance, achieving high efficiency while maintaining high segmentation accuracy. Without bells and whistles, the proposed method, namely DynaMask, brings consistent and noticeable performance improvements over other state-of-the-arts at a moderate computation overhead. The source code: https://github.com/lslrh/DynaMask.

cs.CV↗

arXiv2023

SIM: Semantic-aware Instance Mask Generation for Box-Supervised Instance Segmentation

Weakly supervised instance segmentation using only bounding box annotations has recently attracted much research attention. Most of the current efforts leverage low-level image features as extra supervision without explicitly exploiting the high-level semantic information of the objects, which will become ineffective when the foreground objects have similar appearances to the background or other objects nearby. We propose a new box-supervised instance segmentation approach by developing a Semantic-aware Instance Mask (SIM) generation paradigm. Instead of heavily relying on local pair-wise affinities among neighboring pixels, we construct a group of category-wise feature centroids as prototypes to identify foreground objects and assign them semantic-level pseudo labels. Considering that the semantic-aware prototypes cannot distinguish different instances of the same semantics, we propose a self-correction mechanism to rectify the falsely activated regions while enhancing the correct ones. Furthermore, to handle the occlusions between objects, we tailor the Copy-Paste operation for the weakly-supervised instance segmentation task to augment challenging training data. Extensive experimental results demonstrate the superiority of our proposed SIM approach over other state-of-the-art methods. The source code: https://github.com/lslrh/SIM.

cs.CV↗

arXiv2023

Interactions between a central bubble and a surrounding bubble cluster

The interaction of multiple bubbles is a complex physical problem. A simplified case of multiple bubbles is studied theoretically with a bubble located at the center of a circular bubble cluster. All bubbles in the cluster are equally spaced and own the same initial conditions as the central bubble. The unified theory for bubble dynamics (Zhang et al. arXiv:2301.13698) is applied to model the interaction between the central bubble and the circular bubble cluster. To account for the effect of the propagation time of pressure waves, the emission source of the wave is obtained by interpolating the physical information on the time axis. An underwater explosion experiment with two bubbles of different scales is used to validate the theoretical model. The effect of the bubble cluster with a variation in scale on the pulsation characteristics of the central bubble is studied.

physics.flu-dyn↗

arXiv2023

Improved Regret Bounds for Linear Adversarial MDPs via Linear Optimization

Learning Markov decision processes (MDP) in an adversarial environment has been a challenging problem. The problem becomes even more challenging with function approximation, since the underlying structure of the loss function and transition kernel are especially hard to estimate in a varying environment. In fact, the state-of-the-art results for linear adversarial MDP achieve a regret of $\tilde{O}(K^{6/7})$ ($K$ denotes the number of episodes), which admits a large room for improvement. In this paper, we investigate the problem with a new view, which reduces linear MDP into linear optimization by subtly setting the feature maps of the bandit arms of linear optimization. This new technique, under an exploratory assumption, yields an improved bound of $\tilde{O}(K^{4/5})$ for linear adversarial MDP without access to a transition simulator. The new view could be of independent interest for solving other MDP problems that possess a linear structure.

cs.LG↗

arXiv2023

Dissipationless Layertronics in Axion Insulator $\rm{MnBi_2Te_4}$

Surface electrons in axion insulators are endowed with a topological layer degree of freedom followed by exotic transport phenomena, e.g., the layer Hall effect [Gao et al., Nature 595, 521 (2021)]. Here, we propose that such a layer degree of freedom can be manipulated in a dissipationless way based on the antiferromagnetic $\rm{MnBi_2Te_4}$ with tailored domain structure. This makes $\rm{MnBi_2Te_4}$ a versatile platform to exploit the "layertronics" to encode, process, and store information. Importantly, the layer filter, layer valve, and layer reverser devices can be achieved using the layer-locked chiral domain wall modes. The dissipationless nature of the domain wall modes makes the performance of the layertronic-devices superior to those in spintronics and valleytronics. Specifically, the layer reverser, a layer version of Datta-Das transistor, also fills up the blank in designing the valley reverser in valleytronics. Our work sheds light on constructing new generation electronic devices with high performance and low energy consumption in the framework of layertronics.

cond-mat.mes-hall↗

arXiv2022

Pretraining in Deep Reinforcement Learning: A Survey

The past few years have seen rapid progress in combining reinforcement learning (RL) with deep learning. Various breakthroughs ranging from games to robotics have spurred the interest in designing sophisticated RL algorithms and systems. However, the prevailing workflow in RL is to learn tabula rasa, which may incur computational inefficiency. This precludes continuous deployment of RL algorithms and potentially excludes researchers without large-scale computing resources. In many other areas of machine learning, the pretraining paradigm has shown to be effective in acquiring transferable knowledge, which can be utilized for a variety of downstream tasks. Recently, we saw a surge of interest in Pretraining for Deep RL with promising results. However, much of the research has been based on different experimental settings. Due to the nature of RL, pretraining in this field is faced with unique challenges and hence requires new design principles. In this survey, we seek to systematically review existing works in pretraining for deep reinforcement learning, provide a taxonomy of these methods, discuss each sub-field, and bring attention to open problems and future directions.

cs.LG↗

arXiv2022

Femtosecond intramolecular rearrangement of the CH3NCS radical cation

Strong-field ionization, involving tunnel ionization and electron rescattering, enables femtosecond time-resolved dynamics measurements of chemical reactions involving radical cations. Here, we compare the formation of CH3S+ following the strong-field ionization of the isomers CH3SCN and CH3NCS. The former involves the release of neutral CN, while the latter involves an intramolecular rearrangement. We find the intramolecular rearrangement takes place on the single picosecond timescale and exhibits vibrational coherence. Density functional theory and coupled-cluster calculations on the neutral and singly ionized species help us determine the driving force responsible for intramolecular rearrangement in CH3NCS. Our findings illustrate the complexity that accompanies radical cation chemistry following electron ionization and demonstrate a useful tool for understanding the cation dynamics after ionization.

physics.chem-ph↗

arXiv2022

Salient Object Detection via Dynamic Scale Routing

Recent research advances in salient object detection (SOD) could largely be attributed to ever-stronger multi-scale feature representation empowered by the deep learning technologies. The existing SOD deep models extract multi-scale features via the off-the-shelf encoders and combine them smartly via various delicate decoders. However, the kernel sizes in this commonly-used thread are usually "fixed". In our new experiments, we have observed that kernels of small size are preferable in scenarios containing tiny salient objects. In contrast, large kernel sizes could perform better for images with large salient objects. Inspired by this observation, we advocate the "dynamic" scale routing (as a brand-new idea) in this paper. It will result in a generic plug-in that could directly fit the existing feature backbone. This paper's key technical innovations are two-fold. First, instead of using the vanilla convolution with fixed kernel sizes for the encoder design, we propose the dynamic pyramid convolution (DPConv), which dynamically selects the best-suited kernel sizes w.r.t. the given input. Second, we provide a self-adaptive bidirectional decoder design to accommodate the DPConv-based encoder best. The most significant highlight is its capability of routing between feature scales and their dynamic collection, making the inference process scale-aware. As a result, this paper continues to enhance the current SOTA performance. Both the code and dataset are publicly available at https://github.com/wuzhenyubuaa/DPNet.

cs.CV↗

arXiv2022

Non-equilibrium dynamics of ultracold lattice bosons inside a cavity

We study the non-equilibrium quench dynamics crossing a continuous phase transition between the charge density wave (CDW) and supersolid (SS) phases of a bosonic lattice gas with cavity-mediated interactions. When changing the hopping amplitude in the Hamiltonian as a function of time, we investigate the scaling behavior of the correlation length and vortex density with respect to the quench time and find that there is a threshold of the quench rate separating two distinct scaling regimes. When slowly varying the system below that threshold, we find a power-law scaling as predicted by the Kibble-Zurek mechanism (KZM). While considering fast quench above that threshold, a deviation from the KZM prediction occurs, manifested by a saturation of the defect density. We further show that such distinct scaling behaviors during different dynamic procedures can be understood through comparing the relaxation time and the quench rate.

cond-mat.quant-gas↗

arXiv2022

Hierarchical Conversational Preference Elicitation with Bandit Feedback

The recent advances of conversational recommendations provide a promising way to efficiently elicit users' preferences via conversational interactions. To achieve this, the recommender system conducts conversations with users, asking their preferences for different items or item categories. Most existing conversational recommender systems for cold-start users utilize a multi-armed bandit framework to learn users' preference in an online manner. However, they rely on a pre-defined conversation frequency for asking about item categories instead of individual items, which may incur excessive conversational interactions that hurt user experience. To enable more flexible questioning about key-terms, we formulate a new conversational bandit problem that allows the recommender system to choose either a key-term or an item to recommend at each round and explicitly models the rewards of these actions. This motivates us to handle a new exploration-exploitation (EE) trade-off between key-term asking and item recommendation, which requires us to accurately model the relationship between key-term and item rewards. We conduct a survey and analyze a real-world dataset to find that, unlike assumptions made in prior works, key-term rewards are mainly affected by rewards of representative items. We propose two bandit algorithms, Hier-UCB and Hier-LinUCB, that leverage this observed relationship and the hierarchical structure between key-terms and items to efficiently learn which items to recommend. We theoretically prove that our algorithm can reduce the regret bound's dependency on the total number of items from previous work. We validate our proposed algorithms and regret bound on both synthetic and real-world data.

cs.IR↗

arXiv2022

Characterization and modeling of spiking and bursting in experimental NbOx neuron

Hardware spiking neural networks hold the promise of realizing artificial intelligence with high energy efficiency. In this context, solid-state and scalable memristors can be used to mimic biological neuron characteristics. However, these devices show limited neuronal behaviors and have to be integrated in more complex circuits to implement the rich dynamics of biological neurons. Here we studied a NbOx memristor neuron that is capable of emulating numerous neuronal dynamics, including tonic spiking, stochastic spiking, leaky-integrate-and-fire features, spike latency, temporal integration. The device also exhibits phasic bursting, a property that has scarcely been observed and studied in solid-state nano-neurons. We show that we can reproduce and understand this particular response through simulations using non-linear dynamics. These results show that a single NbOx device is sufficient to emulate a collection of rich neuronal dynamics that paves a path forward for realizing scalable and energy-efficient neuromorphic computing paradigms.

physics.app-ph↗

arXiv2022

Federated Online Clustering of Bandits

Contextual multi-armed bandit (MAB) is an important sequential decision-making problem in recommendation systems. A line of works, called the clustering of bandits (CLUB), utilize the collaborative effect over users and dramatically improve the recommendation quality. Owing to the increasing application scale and public concerns about privacy, there is a growing demand to keep user data decentralized and push bandit learning to the local server side. Existing CLUB algorithms, however, are designed under the centralized setting where data are available at a central server. We focus on studying the federated online clustering of bandit (FCLUB) problem, which aims to minimize the total regret while satisfying privacy and communication considerations. We design a new phase-based scheme for cluster detection and a novel asynchronous communication protocol for cooperative bandit learning for this problem. To protect users' privacy, previous differential privacy (DP) definitions are not very suitable, and we propose a new DP notion that acts on the user cluster level. We provide rigorous proofs to show that our algorithm simultaneously achieves (clustered) DP, sublinear communication complexity and sublinear regret. Finally, experimental evaluations show our superior performance compared with benchmark algorithms.

cs.LG↗

arXiv2022

A Graph-Enhanced Click Model for Web Search

To better exploit search logs and model users' behavior patterns, numerous click models are proposed to extract users' implicit interaction feedback. Most traditional click models are based on the probabilistic graphical model (PGM) framework, which requires manually designed dependencies and may oversimplify user behaviors. Recently, methods based on neural networks are proposed to improve the prediction accuracy of user behaviors by enhancing the expressive ability and allowing flexible dependencies. However, they still suffer from the data sparsity and cold-start problems. In this paper, we propose a novel graph-enhanced click model (GraphCM) for web search. Firstly, we regard each query or document as a vertex, and propose novel homogeneous graph construction methods for queries and documents respectively, to fully exploit both intra-session and inter-session information for the sparsity and cold-start problems. Secondly, following the examination hypothesis, we separately model the attractiveness estimator and examination predictor to output the attractiveness scores and examination probabilities, where graph neural networks and neighbor interaction techniques are applied to extract the auxiliary information encoded in the pre-constructed homogeneous graphs. Finally, we apply combination functions to integrate examination probabilities and attractiveness scores into click predictions. Extensive experiments conducted on three real-world session datasets show that GraphCM not only outperforms the state-of-art models, but also achieves superior performance in addressing the data sparsity and cold-start problems.

cs.IR↗

arXiv2022

Simultaneously Learning Stochastic and Adversarial Bandits with General Graph Feedback

The problem of online learning with graph feedback has been extensively studied in the literature due to its generality and potential to model various learning tasks. Existing works mainly study the adversarial and stochastic feedback separately. If the prior knowledge of the feedback mechanism is unavailable or wrong, such specially designed algorithms could suffer great loss. To avoid this problem, \citet{erez2021towards} try to optimize for both environments. However, they assume the feedback graphs are undirected and each vertex has a self-loop, which compromises the generality of the framework and may not be satisfied in applications. With a general feedback graph, the observation of an arm may not be available when this arm is pulled, which makes the exploration more expensive and the algorithms more challenging to perform optimally in both environments. In this work, we overcome this difficulty by a new trade-off mechanism with a carefully-designed proportion for exploration and exploitation. We prove the proposed algorithm simultaneously achieves $\mathrm{poly} \log T$ regret in the stochastic setting and minimax-optimal regret of $\tilde{O}(T^{2/3})$ in the adversarial setting where $T$ is the horizon and $\tilde{O}$ hides parameters independent of $T$ as well as logarithmic terms. To our knowledge, this is the first best-of-both-worlds result for general feedback graphs.

cs.LG↗

arXiv2022

Comparison-based Conversational Recommender System with Relative Bandit Feedback

With the recent advances of conversational recommendations, the recommender system is able to actively and dynamically elicit user preference via conversational interactions. To achieve this, the system periodically queries users' preference on attributes and collects their feedback. However, most existing conversational recommender systems only enable the user to provide absolute feedback to the attributes. In practice, the absolute feedback is usually limited, as the users tend to provide biased feedback when expressing the preference. Instead, the user is often more inclined to express comparative preferences, since user preferences are inherently relative. To enable users to provide comparative preferences during conversational interactions, we propose a novel comparison-based conversational recommender system. The relative feedback, though more practical, is not easy to be incorporated since its feedback scale is always mismatched with users' absolute preferences. With effectively collecting and understanding the relative feedback from an interactive manner, we further propose a new bandit algorithm, which we call RelativeConUCB. The experiments on both synthetic and real-world datasets validate the advantage of our proposed method, compared to the existing bandit algorithms in the conversational recommender systems.

cs.IR↗

arXiv2022

A novel robot calibration method with plane constraint based on dial indicator

In pace with the electronic technology development and the production technology improvement, industrial robot Give Scope to the Advantage in social services and industrial production. However, due to long-term mechanical wear and structural deformation, the absolute positioning accuracy is low, which greatly hinders the development of manufacturing industry. Calibrating the kinematic parameters of the robot is an effective way to address it. However, the main measuring equipment such as laser trackers and coordinate measuring machines are expensive and need special personnel to operate. Additionally, in the measurement process, due to the influence of many environmental factors, measurement noises are generated, which will affect the calibration accuracy of the robot. Basing on these, we have done the following work: a) developing a robot calibration method based on plane constraint to simplify measurement steps; b) employing Square-root Culture Kalman Filter (SCKF) algorithm for reducing the influence of measurement noises; c) proposing a novel algorithm for identifying kinematic parameters based on SCKF algorithm and Levenberg Marquardt (LM) algorithm to achieve the high calibration accuracy; d) adopting the dial indicator as the measuring equipment for slashing costs. The enough experiments verify the effectiveness of the proposed calibration algorithm and experimental platform.

cs.RO↗