Search

The increasing popularity of deep learning models has created new opportunities for developing AI-based recommender systems. Designing recommender systems using deep neural networks requires careful architecture design, and further optimization demands extensive co-design efforts on jointly optimizing model architecture and hardware. Design automation, such as Automated Machine Learning (AutoML), is necessary to fully exploit the potential of recommender model design, including model choices and model-hardware co-design strategies. We introduce a novel paradigm that utilizes weight sharing to explore abundant solution spaces. Our paradigm creates a large supernet to search for optimal architectures and co-design strategies to address the challenges of data multi-modality and heterogeneity in the recommendation domain. From a model perspective, the supernet includes a variety of operators, dense connectivity, and dimension search options. From a co-design perspective, it encompasses versatile Processing-In-Memory (PIM) configurations to produce hardware-efficient models. Our solution space's scale, heterogeneity, and complexity pose several challenges, which we address by proposing various techniques for training and evaluating the supernet. Our crafted models show promising results on three Click-Through Rates (CTR) prediction benchmarks, outperforming both manually designed and AutoML-crafted models with state-of-the-art performance when focusing solely on architecture search. From a co-design perspective, we achieve 2x FLOPs efficiency, 1.8x energy efficiency, and 1.5x performance improvements in recommender models., Comment: Accepted in ACM Transactions on Recommender Systems. arXiv admin note: substantial text overlap with arXiv:2207.07187

We introduce ajdmom, a Python package designed for automatically deriving moment formulas for the well-established affine jump diffusion (AJD) processes. ajdmom can produce explicit closed-form expressions for moments or conditional moments of any order, significantly enhancing the usability of AJD models. Additionally, ajdmom can compute partial derivatives of these moments with respect to the model parameters, offering a valuable tool for sensitivity analysis. The package's modular architecture makes it easy for adaptation and extension by researchers. ajdmom is open-source and readily available for installation from GitHub or the Python package index (PyPI).

Addressing critical challenges in Lamb wave resonators, this paper presents the first validation of resonators incorporating sub-wavelength through-holes. Using the A3 mode resonator based on a LiNbO3 single-crystal thin film and operating in the K band as a prominent example, we demonstrate the advantages of the through-hole design. In the absence of additional processing steps, and while maintaining device performance--including operating frequency, electromechanical coupling coefficient, and quality factor--without introducing extra spurious modes, this approach effectively reduces the ineffective suspension area of the piezoelectric LN film, potentially enhancing mechanical and thermal stability. It also standardizes etching distances (and times) across various Lamb wave resonators on a single wafer, facilitating the development of Lamb wave filters. The versatility of the through-hole technique, with relaxed constraints on hole geometry and arrangement, further highlights its significance. Together with the other advantages, these features underscore the transformative potential of through-holes in advancing the practical implementation of Lamb wave resonators and filters., Comment: 9 pages, 10 figures

We develop moment estimators for the parameters of affine stochastic volatility models. We first address the challenge of calculating moments for the models by introducing a recursive equation for deriving closed-form expressions for moments of any order. Consequently, we propose our moment estimators. We then establish a central limit theorem for our estimators and derive the explicit formulas for the asymptotic covariance matrix. Finally, we provide numerical results to validate our method.

Gradient-variation online learning aims to achieve regret guarantees that scale with variations in the gradients of online functions, which has been shown to be crucial for attaining fast convergence in games and robustness in stochastic optimization, hence receiving increased attention. Existing results often require the smoothness condition by imposing a fixed bound on gradient Lipschitzness, which may be unrealistic in practice. Recent efforts in neural network optimization suggest a generalized smoothness condition, allowing smoothness to correlate with gradient norms. In this paper, we systematically study gradient-variation online learning under generalized smoothness. We extend the classic optimistic mirror descent algorithm to derive gradient-variation regret by analyzing stability over the optimization trajectory and exploiting smoothness locally. Then, we explore universal online learning, designing a single algorithm with the optimal gradient-variation regrets for convex and strongly convex functions simultaneously, without requiring prior knowledge of curvature. This algorithm adopts a two-layer structure with a meta-algorithm running over a group of base-learners. To ensure favorable guarantees, we design a new Lipschitz-adaptive meta-algorithm, capable of handling potentially unbounded gradients while ensuring a second-order bound to effectively ensemble the base-learners. Finally, we provide the applications for fast-rate convergence in games and stochastic extended adversarial optimization., Comment: NeurIPS 2024

A standard two-qubit joint measurement is the well-known Bell state measurement (BSM), in which each reduced state (traced out one qubit) is the completely mixed state. Recently, a novel quantum joint measurement named elegant joint measurement (EJM) has been proposed, where the reduced states of the EJM basis have tetrahedral symmetry. In this work, we first suggest a five-parameter entangled state and reveal its inherent symmetry. Based on this, we define a more generalized EJM parameterized by $z$, $\varphi$ and $\theta$, and provide the quantum circuits for preparing and detecting these basis states. There are three main results: (i) the previous single-parameter EJM can be directly obtained by specifying the parameters $z$ and $\varphi$; (ii) the initial unit vectors related to the four vertices of the regular tetrahedron are not limited to the original choice and not all the unit vectors in cylindrical coordinates are suitable for forming the EJM basis; and (iii) the reduced states of the present EJM basis can always form two mirrorimage tetrahedrons, robustly preserving its elegant properties. We focus on figuring out what kind of states the EJM basis belongs to and providing a method for constructing the more generalized three-parameter EJM, which may contribute to the multi-setting measurement and the potential applications for quantum information processing., Comment: 9 pages, 3 figures

Foundation models hold significant potential for enabling robots to perform long-horizon general manipulation tasks. However, the simplicity of tasks and the uniformity of environments in existing benchmarks restrict their effective deployment in complex scenarios. To address this limitation, this paper introduces the \textit{RoboCAS} benchmark, the first benchmark specifically designed for complex object arrangement scenarios in robotic manipulation. This benchmark employs flexible and concise scripted policies to efficiently collect a diverse array of demonstrations, showcasing scattered, orderly, and stacked object arrangements within a highly realistic physical simulation environment. It includes complex processes such as target retrieval, obstacle clearance, and robot manipulation, testing agents' abilities to perform long-horizon planning for spatial reasoning and predicting chain reactions under ambiguous instructions. Extensive experiments on multiple baseline models reveal their limitations in managing complex object arrangement scenarios, underscoring the urgent need for intelligent agents capable of performing long-horizon operations in practical deployments and providing valuable insights for future research directions. Project website: \url{https://github.com/notFoundThisPerson/RoboCAS-v0}.

Embodied AI robots have the potential to fundamentally improve the way human beings live and manufacture. Continued progress in the burgeoning field of using large language models to control robots depends critically on an efficient computing substrate. In particular, today's computing systems for embodied AI robots are designed purely based on the interest of algorithm developers, where robot actions are divided into a discrete frame-basis. Such an execution pipeline creates high latency and energy consumption. This paper proposes Corki, an algorithm-architecture co-design framework for real-time embodied AI robot control. Our idea is to decouple LLM inference, robotic control and data communication in the embodied AI robots compute pipeline. Instead of predicting action for one single frame, Corki predicts the trajectory for the near future to reduce the frequency of LLM inference. The algorithm is coupled with a hardware that accelerates transforming trajectory into actual torque signals used to control robots and an execution pipeline that parallels data communication with computation. Corki largely reduces LLM inference frequency by up to 8.0x, resulting in up to 3.6x speed up. The success rate improvement can be up to 17.3%. Code is provided for re-implementation. https://github.com/hyy0613/Corki

Multimodal Large Language Models (MLLM) have made significant progress in the field of document analysis. Despite this, existing benchmarks typically focus only on extracting text and simple layout information, neglecting the complex interactions between elements in structured documents such as mind maps and flowcharts. To address this issue, we introduce the new benchmark named MindBench, which not only includes meticulously constructed bilingual authentic or synthetic images, detailed annotations, evaluation metrics and baseline models, but also specifically designs five types of structured understanding and parsing tasks. These tasks include full parsing, partial parsing, position-related parsing, structured Visual Question Answering (VQA), and position-related VQA, covering key areas such as text recognition, spatial awareness, relationship discernment, and structured parsing. Extensive experimental results demonstrate the substantial potential and significant room for improvement in current models' ability to handle structured document information. We anticipate that the launch of MindBench will significantly advance research and application development in structured document analysis technology. MindBench is available at: https://miasanlei.github.io/MindBench.github.io/., Comment: technical report

Utilizing Vision-Language Models (VLMs) for robotic manipulation represents a novel paradigm, aiming to enhance the model's ability to generalize to new objects and instructions. However, due to variations in camera specifications and mounting positions, existing methods exhibit significant performance disparities across different robotic platforms. To address this challenge, we propose RoboUniView in this paper, an innovative approach that decouples visual feature extraction from action learning. We first learn a unified view representation from multi-perspective views by pre-training on readily accessible data, and then derive actions from this unified view representation to control robotic manipulation. This unified view representation more accurately mirrors the physical world and is not constrained by the robotic platform's camera parameters. Thanks to this methodology, we achieve state-of-the-art performance on the demanding CALVIN benchmark, enhancing the success rate in the $D \to D$ setting from 93.0% to 96.2%, and in the $ABC \to D$ setting from 92.2% to 94.2%. Moreover, our model exhibits outstanding adaptability and flexibility: it maintains high performance under unseen camera parameters, can utilize multiple datasets with varying camera parameters, and is capable of joint cross-task learning across datasets. Code is provided for re-implementation. https://github.com/liufanfanlff/RoboUniview

Despite extensive exploration of acoustic vortices carrying orbital angular momentum (OAM), the generation of acoustic vortices with OAM orientations beyond the conventional longitudinal direction remains largely unexplored. Spatiotemporal (ST) vortices, featuring spiral phase twisting in the ST domain and carrying transverse OAM, have recently attracted considerable interest in optics and acoustics. Here, we report the generation of three-dimensional (3D) ST acoustic vortices with arbitrarily oriented OAM, thereby opening up a new dimension in acoustic OAM control. By utilizing a two-dimensional (2D) acoustic phased array, we introduce two approaches to manipulate the orientation of OAM: through the direct rotation of vortices in 3D space and the intersection of vortices carrying distinct types of OAM. These methods enable unprecedented control over the orientation of acoustic OAM, providing a new degree of freedom in the manipulation of acoustic waves. The arbitrarily oriented OAM holds promise for enhancing acoustic communication by broadening capacity and enabling more complex particle manipulation techniques. Our work establishes a foundation for future explorations into the complex dynamics of novel structured acoustic fields in the ST domain.

We study the mechanism of topological mass-generation for 3d Chern-Simons gauge theories and propose a brand-new Topological Equivalence Theorem to connect scattering amplitudes of the physical gauge boson states to that of the transverse states under high energy expansion. We prove a general energy cancellation mechanism for $N$-point physical gauge boson amplitudes, which predicts large cancellations of $E^{4-L}\to E^{(4-L)- N}$ at any $L$-loop level ($L\geqslant 0$). We extend the double-copy approach to construct massive graviton amplitudes and study their structures. We newly uncover a series of strikingly large energy cancellations $E^{12}\to E^1$ of the tree-level four-graviton scattering amplitude under high energy expansion and establish a new correspondence between the two energy cancellations in the topologically massive Yang-Mills gauge theory and the topologically massive gravity theory. We further study the scattering amplitudes of Chern-Simons gauge bosons and gravitons in the nonrelativistic limit., Comment: 8+5 pages, to match Journal Version. This Letter paper is a companion of the systematic longer paper arXiv:2110.05399 (46pp)

This work proposes a double-layer thin-film lithium niobate (LiNbO3) longitudinally excited shear wave resonator with a theoretical electromechanical coupling coefficient exceeding 60%, RaR close to 28%, and no spurious modes. This ultra-large electromechanical coupling coefficient, which is close to the upper limit of LiNbO3, is much larger than all microwave acoustic resonators reported so far. Based on X-cut thin-film LiNbO3, when the film thickness is in the order of hundreds of nanometers, the frequency of the fundamental mode of the resonator can cover 1GHz to10GHz. The resonator design is convenient and flexible. The resonant frequency can be modulated monotonically by changing either the electrode or the thickness of the thin-film LiNbO3 without introducing additional spurious modes. This ideal resonator architecture is also applicable to LiTaO3. With the development of the new generation of mobile communications, this resonator is expected to become a key solution for future high-performance, ultra-wide-bandwidth acoustic filters., Comment: 15 pages,9 figures

Background and Objective: In neurosurgery, fusing clinical images and depth images that can improve the information and details is beneficial to surgery. We found that the registration of face depth images was invalid frequently using existing methods. To abundant traditional image methods with depth information, a method in registering with depth images and traditional clinical images was investigated. Methods: We used the dlib library, a C++ library that could be used in face recognition, and recognized the key points on faces from the structure light camera and CT image. The two key point clouds were registered for coarse registration by the ICP method. Fine registration was finished after coarse registration by the ICP method. Results: RMSE after coarse and fine registration is as low as 0.995913 mm. Compared with traditional methods, it also takes less time. Conclusions: The new method successfully registered the facial depth image from structure light images and CT with a low error, and that would be promising and efficient in clinical application of neurosurgery.

Decentralized Multi-agent Learning (DML) enables collaborative model training while preserving data privacy. However, inherent heterogeneity in agents' resources (computation, communication, and task size) may lead to substantial variations in training time. This heterogeneity creates a bottleneck, lengthening the overall training time due to straggler effects and potentially wasting spare resources of faster agents. To minimize training time in heterogeneous environments, we present a Communication-Efficient Training Workload Balancing for Decentralized Multi-Agent Learning (ComDML), which balances the workload among agents through a decentralized approach. Leveraging local-loss split training, ComDML enables parallel updates, where slower agents offload part of their workload to faster agents. To minimize the overall training time, ComDML optimizes the workload balancing by jointly considering the communication and computation capacities of agents, which hinges upon integer programming. A dynamic decentralized pairing scheduler is developed to efficiently pair agents and determine optimal offloading amounts. We prove that in ComDML, both slower and faster agents' models converge, for convex and non-convex functions. Furthermore, extensive experimental results on popular datasets (CIFAR-10, CIFAR-100, and CINIC-10) and their non-I.I.D. variants, with large models such as ResNet-56 and ResNet-110, demonstrate that ComDML can significantly reduce the overall training time while maintaining model accuracy, compared to state-of-the-art methods. ComDML demonstrates robustness in heterogeneous environments, and privacy measures can be seamlessly integrated for enhanced data protection., Comment: This paper has been accepted for presentation at ICDCS (44th IEEE International Conference on Distributed Computing Systems). Keywords: decentralized multi-agent learning, federated learning, edge computing, heterogeneous agents, workload balancing, and communication-efficient training )

Exploring dense connectivity of convolutional operators establishes critical "synapses" to communicate feature vectors from different levels and enriches the set of transformations on Computer Vision applications. Yet, even with heavy-machinery approaches such as Neural Architecture Search (NAS), discovering effective connectivity patterns requires tremendous efforts due to either constrained connectivity design space or a sub-optimal exploration process induced by an unconstrained search space. In this paper, we propose CSCO, a novel paradigm that fabricates effective connectivity of convolutional operators with minimal utilization of existing design motifs and further utilizes the discovered wiring to construct high-performing ConvNets. CSCO guides the exploration via a neural predictor as a surrogate of the ground-truth performance. We introduce Graph Isomorphism as data augmentation to improve sample efficiency and propose a Metropolis-Hastings Evolutionary Search (MH-ES) to evade locally optimal architectures and advance search quality. Results on ImageNet show ~0.6% performance improvement over hand-crafted and NAS-crafted dense connectivity. Our code is publicly available., Comment: To appear on Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops (2024)

First-come first-serve scheduling can result in substantial (up to 10%) of transiently idle nodes on supercomputers. Recognizing that such unfilled nodes are well-suited for deep neural network (DNN) training, due to the flexible nature of DNN training tasks, Liu et al. proposed that the re-scaling DNN training tasks to fit gaps in schedules be formulated as a mixed-integer linear programming (MILP) problem, and demonstrated via simulation the potential benefits of the approach. Here, we introduce MalleTrain, a system that provides the first practical implementation of this approach and that furthermore generalizes it by allowing it use even for DNN training applications for which model information is unknown before runtime. Key to this latter innovation is the use of a lightweight online job profiling advisor (JPA) to collect critical scalability information for DNN jobs -- information that it then employs to optimize resource allocations dynamically, in real time. We describe the MalleTrain architecture and present the results of a detailed experimental evaluation on a supercomputer GPU cluster and several representative DNN training workloads, including neural architecture search and hyperparameter optimization. Our results not only confirm the practical feasibility of leveraging idle supercomputer nodes for DNN training but improve significantly on prior results, improving training throughput by up to 22.3\% without requiring users to provide job scalability information.

A quantum network shared entangled sources among distant nodes enables us to distribute entanglement along the network by suitable measurements. Network nonlocality means that it does not admit a network model involving local variables emitted from independent sources. In this work, we construct an $(n,m,p)$-type quantum network configuration and then derive the corresponding $n$-local correlation inequalities based on the assumption of independent sources. As a universal acyclic network configuration, it can cover most of the existing network models, such as the typical chain-network and star-network, and admit both centerless and asymmetric configurations. Then we demonstrate the non-$n$-locality of the present network by calculating the violation of the $n$-local inequality with bipartite entangled sources and Pauli measurements., Comment: 7 pages, 2 figures

Spin and orbital angular momenta are fundamental physical characteristics described by polarization and spatial degrees of freedom, respectively. Polarization is a feature of vector fields while spatial phase gradient determines the orbital angular momentum ubiquitous to any scalar field. Common wisdom treats these two degrees of freedom as distinct and independent principles to manipulate wave propagations. Here, we demonstrate their synergy. This is achieved by introducing two orthogonal $p$-orbitals as eigenbases, whose spatial modal features are exploited to generate orbital angular momenta and the associated orbital orientations provide means to simultaneously manipulate polarizations. Through periodic modulation and directional coupling, we realize a full cyclic evolution of the synchronized and synergized spin-orbital angular momenta. Remarkably, this evolution acquires a nontrivial geometric phase, leading to its representation on a M\"obius strip. Experimentally, an acoustic cavity array is designed, whose dipole resonances precisely mimic the $p$-orbitals. The acoustic waves, uniquely, see the pressure (scalar) field as a spatial feature and carry an intrinsic polarization defined by the velocity (vector) field, serving as an ideal platform to observe the synergy of spin and orbital angular momenta. Based on such a property, we further showcase a spin-orbital-Hall effect, highlighting the intricate locking of handedness, directionality, spin density and spatial mode profile. Our study unveils a fundamental connection between spin and orbital angular momenta, promising avenues for novel applications in information coding and high-capacity communications.

The utilization of residual learning has become widespread in deep and scalable neural nets. However, the fundamental principles that contribute to the success of residual learning remain elusive, thus hindering effective training of plain nets with depth scalability. In this paper, we peek behind the curtains of residual learning by uncovering the "dissipating inputs" phenomenon that leads to convergence failure in plain neural nets: the input is gradually compromised through plain layers due to non-linearities, resulting in challenges of learning feature representations. We theoretically demonstrate how plain neural nets degenerate the input to random noise and emphasize the significance of a residual connection that maintains a better lower bound of surviving neurons as a solution. With our theoretical discoveries, we propose "The Plain Neural Net Hypothesis" (PNNH) that identifies the internal path across non-linear layers as the most critical part in residual learning, and establishes a paradigm to support the training of deep plain neural nets devoid of residual connections. We thoroughly evaluate PNNH-enabled CNN architectures and Transformers on popular vision benchmarks, showing on-par accuracy, up to 0.3% higher training throughput, and 2x better parameter efficiency compared to ResNets and vision Transformers., Comment: Arxiv Preprint

Spinal metastasis is the most common disease in bone metastasis and may cause pain, instability and neurological injuries. Early detection of spinal metastasis is critical for accurate staging and optimal treatment. The diagnosis is usually facilitated with Computed Tomography (CT) scans, which requires considerable efforts from well-trained radiologists. In this paper, we explore a learning-based automatic bone quality classification method for spinal metastasis based on CT images. We simultaneously take the posterolateral spine involvement classification task into account, and employ multi-task learning (MTL) technique to improve the performance. MTL acts as a form of inductive bias which helps the model generalize better on each task by sharing representations between related tasks. Based on the prior knowledge that the mixed type can be viewed as both blastic and lytic, we model the task of bone quality classification as two binary classification sub-tasks, i.e., whether blastic and whether lytic, and leverage a multiple layer perceptron to combine their predictions. In order to make the model more robust and generalize better, self-paced learning is adopted to gradually involve from easy to more complex samples into the training process. The proposed learning-based method is evaluated on a proprietary spinal metastasis CT dataset. At slice level, our method significantly outperforms an 121-layer DenseNet classifier in sensitivities by $+12.54\%$, $+7.23\%$ and $+29.06\%$ for blastic, mixed and lytic lesions, respectively, meanwhile $+12.33\%$, $+23.21\%$ and $+34.25\%$ at vertebrae level.

Vertebral body (VB) segmentation is an important preliminary step towards medical visual diagnosis for spinal diseases. However, most previous works require pixel/voxel-wise strong supervisions, which is expensive, tedious and time-consuming for experts to annotate. In this paper, we propose a Weakly supervised Iterative Spinal Segmentation (WISS) method leveraging only four corner landmark weak labels on a single sagittal slice to achieve automatic volumetric segmentation from CT images for VBs. WISS first segments VBs on an annotated sagittal slice in an iterative self-training manner. This self-training method alternates between training and refining labels in the training set. Then WISS proceeds to segment the whole VBs slice by slice with a slice-propagation method to obtain volumetric segmentations. We evaluate the performance of WISS on a private spinal metastases CT dataset and the public lumbar CT dataset. On the first dataset, WISS achieves distinct improvements with regard to two different backbones. For the second dataset, WISS achieves dice coefficients of $91.7\%$ and $83.7\%$ for mid-sagittal slices and 3D CT volumes, respectively, saving a lot of labeling costs and only sacrificing a little segmentation performance., Comment: arXiv admin note: text overlap with arXiv:1412.7062 by other authors

The charge density wave (CDW), as a hallmark of vanadium-based kagome superconductor AV3Sb5 (A = K, Rb, Cs), has attracted intensive attention. However, the fundamental controversy regarding the underlying mechanism of CDW therein persists. Recently, the vanadium-based bi-layered kagome metal ScV6Sn6, reported to exhibit a long-range charge order below 94 K, has emerged as a promising candidate to further clarify this core issue. Here, employing micro-focusing angle-resolved photoemission spectroscopy ({\mu}-ARPES) and first-principles calculations, we systematically studied the unique CDW order in vanadium-based bi-layered kagome metals by comparing ScV6Sn6 with its isostructural counterpart YV6Sn6, which lacks a CDW ground state. Combining ARPES data and the corresponding joint density of states (DOS), we suggest that the VHS nesting mechanism might be invalid in these materials. Besides, in ScV6Sn6, we identified multiple hybridization energy gaps resulting from CDW-induced band folding, along with an anomalous band dispersion, implying a potential electron-phonon coupling driven mechanism underlying the formation of the CDW order. Our finding not only comprehensively maps the electronic structure of V-based bi-layer kagome metals but also provide constructive experimental evidence for the unique origin of CDW in this system., Comment: 14 pages, 5 figures

Surface defect inspection plays an important role in the process of industrial manufacture and production. Though Convolutional Neural Network (CNN) based defect inspection methods have made huge leaps, they still confront a lot of challenges such as defect scale variation, complex background, low contrast, and so on. To address these issues, we propose a joint attention-guided feature fusion network (JAFFNet) for saliency detection of surface defects based on the encoder-decoder network. JAFFNet mainly incorporates a joint attention-guided feature fusion (JAFF) module into decoding stages to adaptively fuse low-level and high-level features. The JAFF module learns to emphasize defect features and suppress background noise during feature fusion, which is beneficial for detecting low-contrast defects. In addition, JAFFNet introduces a dense receptive field (DRF) module following the encoder to capture features with rich context information, which helps detect defects of different scales. The JAFF module mainly utilizes a learned joint channel-spatial attention map provided by high-level semantic features to guide feature fusion. The attention map makes the model pay more attention to defect features. The DRF module utilizes a sequence of multi-receptive-field (MRF) units with each taking as inputs all the preceding MRF feature maps and the original input. The obtained DRF features capture rich context information with a large range of receptive fields. Extensive experiments conducted on SD-saliency-900, Magnetic tile, and DAGM 2007 indicate that our method achieves promising performance in comparison with other state-of-the-art methods. Meanwhile, our method reaches a real-time defect detection speed of 66 FPS.

As a generalization of the well-known Bell state measurement (BSM), the elegant joint measurement (EJM) is a kind of novel two-qubit joint measurement, parameterized by a subtle phase factor $\theta \in [0,\pi/2]$. We explore quantum teleportation based on the EJM, inspired by Gisin's idea that quantum entanglement not only provides quantum channel and also quantum joint measurement for quantum teleportation. It is a probabilistic teleportation caused by undesired nonunitary quantum evolution. There are two interesting features in the present scenario. First, it goes beyond the conventional teleportation scenario, which can be included in the present scenario. Second, different from the BSM being single input and four outcomes, it can provide an adjustable input setting or even multiple measurement settings for the sender (or the controller). Moreover, we show in detail the feasible quantum circuits to realize the present scenario, where a few unitary operations and a nonunitary quantum gate are being utilized., Comment: 8 pages, 3 figures

We introduce Grounded SAM, which uses Grounding DINO as an open-set object detector to combine with the segment anything model (SAM). This integration enables the detection and segmentation of any regions based on arbitrary text inputs and opens a door to connecting various vision models. As shown in Fig.1, a wide range of vision tasks can be achieved by using the versatile Grounded SAM pipeline. For example, an automatic annotation pipeline based solely on input images can be realized by incorporating models such as BLIP and Recognize Anything. Additionally, incorporating Stable-Diffusion allows for controllable image editing, while the integration of OSX facilitates promptable 3D human motion analysis. Grounded SAM also shows superior performance on open-vocabulary benchmarks, achieving 48.7 mean AP on SegInW (Segmentation in the wild) zero-shot benchmark with the combination of Grounding DINO-Base and SAM-Huge models.

Federated learning (FL) enables collaboratively training a model while keeping the training data decentralized and private. However, one significant impediment to training a model using FL, especially large models, is the resource constraints of devices with heterogeneous computation and communication capacities as well as varying task sizes. Such heterogeneity would render significant variations in the training time of clients, resulting in a longer overall training time as well as a waste of resources in faster clients. To tackle these heterogeneity issues, we propose the Dynamic Tiering-based Federated Learning (DTFL) system where slower clients dynamically offload part of the model to the server to alleviate resource constraints and speed up training. By leveraging the concept of Split Learning, DTFL offloads different portions of the global model to clients in different tiers and enables each client to update the models in parallel via local-loss-based training. This helps reduce the computation and communication demand on resource-constrained devices and thus mitigates the straggler problem. DTFL introduces a dynamic tier scheduler that uses tier profiling to estimate the expected training time of each client, based on their historical training time, communication speed, and dataset size. The dynamic tier scheduler assigns clients to suitable tiers to minimize the overall training time in each round. We first theoretically prove the convergence properties of DTFL. We then train large models (ResNet-56 and ResNet-110) on popular image datasets (CIFAR-10, CIFAR-100, CINIC-10, and HAM10000) under both IID and non-IID systems. Extensive experimental results show that compared with state-of-the-art FL methods, DTFL can significantly reduce the training time while maintaining model accuracy.

Search efficiency and serving efficiency are two major axes in building feature interactions and expediting the model development process in recommender systems. On large-scale benchmarks, searching for the optimal feature interaction design requires extensive cost due to the sequential workflow on the large volume of data. In addition, fusing interactions of various sources, orders, and mathematical operations introduces potential conflicts and additional redundancy toward recommender models, leading to sub-optimal trade-offs in performance and serving cost. In this paper, we present DistDNAS as a neat solution to brew swift and efficient feature interaction design. DistDNAS proposes a supernet to incorporate interaction modules of varying orders and types as a search space. To optimize search efficiency, DistDNAS distributes the search and aggregates the choice of optimal interaction modules on varying data dates, achieving over 25x speed-up and reducing search cost from 2 days to 2 hours. To optimize serving efficiency, DistDNAS introduces a differentiable cost-aware loss to penalize the selection of redundant interaction modules, enhancing the efficiency of discovered feature interactions in serving. We extensively evaluate the best models crafted by DistDNAS on a 1TB Criteo Terabyte dataset. Experimental evaluations demonstrate 0.001 AUC improvement and 60% FLOPs saving over current state-of-the-art CTR models.