Search

Effective multi-agent collaboration is imperative for solving complex, distributed problems. In this context, two key challenges must be addressed: first, autonomously identifying optimal objectives for collective outcomes; second, aligning these objectives among agents. Traditional frameworks, often reliant on centralized learning, struggle with scalability and efficiency in large multi-agent systems. To overcome these issues, we introduce a decentralized state-based value learning algorithm that enables agents to independently discover optimal states. Furthermore, we introduce a novel mechanism for multi-agent interaction, wherein less proficient agents follow and adopt policies from more experienced ones, thereby indirectly guiding their learning process. Our theoretical analysis shows that our approach leads decentralized agents to an optimal collective policy. Empirical experiments further demonstrate that our method outperforms existing decentralized state-based and action-based value learning strategies by effectively identifying and aligning optimal objectives., Comment: 10 pages, 3 figures, extended version of our 2024 American Control Conference publication

We study incentivized exploration for the multi-armed bandit (MAB) problem with non-stationary reward distributions, where players receive compensation for exploring arms other than the greedy choice and may provide biased feedback on the reward. We consider two different non-stationary environments: abruptly-changing and continuously-changing, and propose respective incentivized exploration algorithms. We show that the proposed algorithms achieve sublinear regret and compensation over time, thus effectively incentivizing exploration despite the nonstationarity and the biased or drifted feedback.

We introduce NeuV-SLAM, a novel dense simultaneous localization and mapping pipeline based on neural multiresolution voxels, characterized by ultra-fast convergence and incremental expansion capabilities. This pipeline utilizes RGBD images as input to construct multiresolution neural voxels, achieving rapid convergence while maintaining robust incremental scene reconstruction and camera tracking. Central to our methodology is to propose a novel implicit representation, termed VDF that combines the implementation of neural signed distance field (SDF) voxels with an SDF activation strategy. This approach entails the direct optimization of color features and SDF values anchored within the voxels, substantially enhancing the rate of scene convergence. To ensure the acquisition of clear edge delineation, SDF activation is designed, which maintains exemplary scene representation fidelity even under constraints of voxel resolution. Furthermore, in pursuit of advancing rapid incremental expansion with low computational overhead, we developed hashMV, a novel hash-based multiresolution voxel management structure. This architecture is complemented by a strategically designed voxel generation technique that synergizes with a two-dimensional scene prior. Our empirical evaluations, conducted on the Replica and ScanNet Datasets, substantiate NeuV-SLAM's exceptional efficacy in terms of convergence speed, tracking accuracy, scene reconstruction, and rendering quality.

Cross-lingual entity alignment is the task of finding the same semantic entities from different language knowledge graphs. In this paper, we propose a simple and novel unsupervised method for cross-language entity alignment. We utilize the deep learning multi-language encoder combined with a machine translator to encode knowledge graph text, which reduces the reliance on label data. Unlike traditional methods that only emphasize global or local alignment, our method simultaneously considers both alignment strategies. We first view the alignment task as a bipartite matching problem and then adopt the re-exchanging idea to accomplish alignment. Compared with the traditional bipartite matching algorithm that only gives one optimal solution, our algorithm generates ranked matching results which enabled many potentials downstream tasks. Additionally, our method can adapt two different types of optimization (minimal and maximal) in the bipartite matching process, which provides more flexibility. Our evaluation shows, we each scored 0.966, 0.990, and 0.996 Hits@1 rates on the DBP15K dataset in Chinese, Japanese, and French to English alignment tasks. We outperformed the state-of-the-art method in unsupervised and semi-supervised categories. Compared with the state-of-the-art supervised method, our method outperforms 2.6% and 0.4% in Ja-En and Fr-En alignment tasks while marginally lower by 0.2% in the Zh-En alignment task., Comment: 17 pages,5 figures, Accepted by ECML PKDD 2023(Research Track)

The formation mechanism of high-concentration dwarf galaxies is still a mystery. We perform a comparative study of the intrinsic shape of nearby low-mass galaxies with different stellar concentration. The intrinsic shape is parameterized by the intermediate-to-major axis ratios B/A and the minor-to-major axis ratios C/A of triaxial ellipsoidal models. Our galaxies ($10^{7.5} M_\odot$ < $M_\star$ < $10^{10.0} M_\odot$) are selected to have spectroscopic redshift from SDSS or GAMA, and have broadband optical images from the HSC-SSP Wide layer survey. The deep HSC-SSP images allow to measure the apparent axis ratios $q$ at galactic radii beyond the central star-forming area of our galaxies. We infer the intrinsic axis ratios based on the $q$ distributions. We find that 1) our galaxies have typical intrinsic shape similarly close to be oblate ($\mu_{B/A}$ $\sim$ 0.9--1), regardless of the concentration, stellar mass, star formation activity, and local environment (being central or satellite); 2) galaxies with the highest concentration tend to have intrinsic thickness similar to or (in virtually all cases) slightly thinner (i.e. smaller mean $\mu_{C/A}$ or equivalently lower triaxiality) than ordinary galaxies, regardless of other properties explored here. This appears to be in contrast with the expectation of the classic merger scenario for high-concentration galaxies. Given the lack of a complete understanding of dwarf-dwarf merger, we cannot draw a definite conclusion about the relevance of mergers in the formation of high-concentration dwarfs. Other mechanisms such as halo spin may also play important roles in the formation of high-concentration dwarf galaxies., Comment: 12 pages, 8 figures, 2 tables, accepted for publication in ApJ

Purpose: Congenital heart defect (CHD) is the most common birth defect. Thoracic echocardiography (TTE) can provide sufficient cardiac structure information, evaluate hemodynamics and cardiac function, and is an effective method for atrial septal defect (ASD) examination. This paper aims to study a deep learning method based on cardiac ultrasound video to assist in ASD diagnosis. Materials and methods: We select two standard views of the atrial septum (subAS) and low parasternal four-compartment view (LPS4C) as the two views to identify ASD. We enlist data from 300 children patients as part of a double-blind experiment for five-fold cross-validation to verify the performance of our model. In addition, data from 30 children patients (15 positives and 15 negatives) are collected for clinician testing and compared to our model test results (these 30 samples do not participate in model training). We propose an echocardiography video-based atrial septal defect diagnosis system. In our model, we present a block random selection, maximal agreement decision and frame sampling strategy for training and testing respectively, resNet18 and r3D networks are used to extract the frame features and aggregate them to build a rich video-level representation. Results: We validate our model using our private dataset by five-cross validation. For ASD detection, we achieve 89.33 AUC, 84.95 accuracy, 85.70 sensitivity, 81.51 specificity and 81.99 F1 score. Conclusion: The proposed model is multiple instances learning-based deep learning model for video atrial septal defect detection which effectively improves ASD detection accuracy when compared to the performances of previous networks and clinical doctors.

We report our experiences implementing standards-based grading at scale in an Algorithms course, which serves as the terminal required CS Theory course in our department's undergraduate curriculum. The course had 200-400 students, taught by two instructors, eight graduate teaching assistants, and supported by two additional graders and several undergraduate course assistants. We highlight the role of standards-based grading in supporting our students during the COVID-19 pandemic. We conclude by detailing the successes and adjustments we would make to the course structure., Comment: This is an extended version of our paper, which will appear in ITiCSE 2022

In this paper, we propose an optimal control-estimation architecture for distribution networks, which jointly solves the optimal power flow (OPF) problem and static state estimation (SE) problem through an online gradient-based feedback algorithm. The main objective is to enable a fast and timely interaction between the optimal controllers and state estimators with limited sensor measurements. First, convergence and optimality of the proposed algorithm are analytically established. Then, the proposed gradient-based algorithm is modified by introducing statistical information of the inherent estimation and linearization errors for an improved and robust performance of the online control decisions. Overall, the proposed method eliminates the traditional separation of control and operation, where control and estimation usually operate at distinct layers and different time-scales. Hence, it enables a computationally affordable, efficient and robust online operational framework for distribution networks under time-varying settings.

In this paper, we study the problem of jointly estimating the optical flow and scene flow from synchronized 2D and 3D data. Previous methods either employ a complex pipeline that splits the joint task into independent stages, or fuse 2D and 3D information in an "early-fusion" or "late-fusion" manner. Such one-size-fits-all approaches suffer from a dilemma of failing to fully utilize the characteristic of each modality or to maximize the inter-modality complementarity. To address the problem, we propose a novel end-to-end framework, called CamLiFlow. It consists of 2D and 3D branches with multiple bidirectional connections between them in specific layers. Different from previous work, we apply a point-based 3D branch to better extract the geometric features and design a symmetric learnable operator to fuse dense image features and sparse point features. Experiments show that CamLiFlow achieves better performance with fewer parameters. Our method ranks 1st on the KITTI Scene Flow benchmark, outperforming the previous art with 1/7 parameters. Code is available at https://github.com/MCG-NJU/CamLiFlow., Comment: Accepted to CVPR 2022 (Oral)

Adaptive charging can charge electric vehicles (EVs) at scale cost effectively, despite the uncertainty in EV arrivals. We formulate adaptive EV charging as a feasibility problem that meets all EVs' energy demands before their deadlines while satisfying constraints in charging rate and total charging power. We propose an online algorithm, smoothed least-laxity-first (sLLF), that decides the current charging rates without the knowledge of future arrivals and demands. We characterize the performance of the sLLF algorithm analytically and numerically. Numerical experiments with real-world data show that it has a significantly higher rate of feasible EV charging than several other existing EV charging algorithms. Resource augmentation framework is employed to assess the feasibility condition of the algorithm. The assessment shows that the sLLF algorithm achieves perfect feasibility with only a 0.07 increase in resources., Comment: 14 pages, 4 figures

A key challenge in wide adoption of sophisticated context-aware applications is the requirement of continuous sensing and context computing. This paper presents Panorama, a middleware that identifies collaboration opportunities to offload context computing tasks to nearby mobile devices as well as cloudlets/cloud. At the heart of Panorama is a multi-objective optimizer that takes into account different constraints such as access cost, computation capability, access latency, energy consumption and data privacy, and efficiently computes a collaboration plan optimized simultaneously for different objectives such as minimizing cost, energy and/or execution time. Panorama provides support for discovering nearby devices and cloudlets/cloud, computing an optimal collaboration plan, distributing computation to participating devices, and getting the results back. The paper provides an extensive evaluation of Panorama via two representative context monitoring applications over a set of Android devices and a cloudlet/cloud under different constraints., Comment: Published in MobiCase 2015

The increasing penetration of distributed energy resources (DERs) in the distribution networks has turned the conventionally passive load buses into active buses that can provide grid services for the transmission system. To take advantage of the DERs in the distribution networks, this letter formulates a transmission-and-distribution (T&D) systems co-optimization problem that achieves economic dispatch at the transmission level and optimal voltage regulation at the distribution level by leveraging large generators and DERs. A primal-dual gradient algorithm is proposed to solve this optimization problem jointly for T&D systems, and a distributed market-based equivalent of the gradient algorithm is used for practical implementation. The results are corroborated by numerical examples with the IEEE 39-Bus system connected with 7 different distribution networks.

Multi-robot cooperation requires agents to make decisions that are consistent with the shared goal without disregarding action-specific preferences that might arise from asymmetry in capabilities and individual objectives. To accomplish this goal, we propose a method named SLiCC: Stackelberg Learning in Cooperative Control. SLiCC models the problem as a partially observable stochastic game composed of Stackelberg bimatrix games, and uses deep reinforcement learning to obtain the payoff matrices associated with these games. Appropriate cooperative actions are then selected with the derived Stackelberg equilibria. Using a bi-robot cooperative object transportation problem, we validate the performance of SLiCC against centralized multi-agent Q-learning and demonstrate that SLiCC achieves better combined utility., Comment: 8 pages, 7 figures

We investigate the sparse linear contextual bandit problem where the parameter $\theta$ is sparse. To relieve the sampling inefficiency, we utilize the "perturbed adversary" where the context is generated adversarilly but with small random non-adaptive perturbations. We prove that the simple online Lasso supports sparse linear contextual bandit with regret bound $\mathcal{O}(\sqrt{kT\log d})$ even when $d \gg T$ where $k$ and $d$ are the number of effective and ambient dimension, respectively. Compared to the recent work from Sivakumar et al. (2020), our analysis does not rely on the precondition processing, adaptive perturbation (the adaptive perturbation violates the i.i.d perturbation setting) or truncation on the error set. Moreover, the special structures in our results explicitly characterize how the perturbation affects exploration length, guide the design of perturbation together with the fundamental performance limit of perturbation method. Numerical experiments are provided to complement the theoretical analysis., Comment: 16 pages, 2 figures