亚洲男人的天堂2018av,欧美草比,久久久久久免费视频精选,国色天香在线看免费,久久久久亚洲av成人片仓井空

This work introduces ExaLogLog, a new data structure for approximate distinct counting, which has the same practical properties as the popular HyperLogLog algorithm. It is commutative, idempotent, mergeable, reducible, has a constant-time insert operation, and supports distinct counts up to the exa-scale. At the same time, as theoretically derived and experimentally verified, it requires 43% less space to achieve the same estimation error.

Gradient sparsification is a communication optimisation technique for scaling and accelerating distributed deep neural network (DNN) training. It reduces the increasing communication traffic for gradient aggregation. However, existing sparsifiers have poor scalability because of the high computational cost of gradient selection and/or increase in communication traffic. In particular, an increase in communication traffic is caused by gradient build-up and inappropriate threshold for gradient selection. To address these challenges, we propose a novel gradient sparsification method called MiCRO. In MiCRO, the gradient vector is partitioned, and each partition is assigned to the corresponding worker. Each worker then selects gradients from its partition, and the aggregated gradients are free from gradient build-up. Moreover, MiCRO estimates the accurate threshold to maintain the communication traffic as per user requirement by minimising the compression ratio error. MiCRO enables near-zero cost gradient sparsification by solving existing problems that hinder the scalability and acceleration of distributed DNN training. In our extensive experiments, MiCRO outperformed state-of-the-art sparsifiers with an outstanding convergence rate.

The increasing adoption of solar energy necessitates advanced methodologies for monitoring and maintenance to ensure optimal performance of solar panel installations. A critical component in this context is the accurate segmentation of solar panels from aerial or satellite imagery, which is essential for identifying operational issues and assessing efficiency. This paper addresses the significant challenges in panel segmentation, particularly the scarcity of annotated data and the labour-intensive nature of manual annotation for supervised learning. We explore and apply Self-Supervised Learning (SSL) to solve these challenges. We demonstrate that SSL significantly enhances model generalization under various conditions and reduces dependency on manually annotated data, paving the way for robust and adaptable solar panel segmentation solutions.

There has been an increasing interest in large speech models that can perform multiple speech processing tasks in a single model. Such models usually adopt the encoder-decoder or decoder-only architecture due to their popularity and good performance in many domains. However, autoregressive models can be slower during inference compared to non-autoregressive models and also have potential risks of hallucination. Though prior studies observed promising results of non-autoregressive models for certain tasks at small scales, it remains unclear if they can be scaled to speech-to-text generation in diverse languages and tasks. Inspired by the Open Whisper-style Speech Model (OWSM) project, we propose OWSM-CTC, a novel encoder-only speech foundation model based on Connectionist Temporal Classification (CTC). It is trained on 180k hours of public audio data for multilingual automatic speech recognition (ASR), speech translation (ST), and language identification (LID). Compared to encoder-decoder OWSM, our OWSM-CTC achieves competitive results on ASR and up to 25% relative improvement on ST, while it is more robust and 3 to 4 times faster for inference. OWSM-CTC also improves the long-form ASR result with 20x speed-up. We will publicly release our codebase, pre-trained model, and training logs to promote open science in speech foundation models.

This paper introduces PDEformer, a neural solver for partial differential equations (PDEs) capable of simultaneously addressing various types of PDEs. We advocate representing the PDE in the form of a computational graph, facilitating the seamless integration of both symbolic and numerical information inherent in a PDE. A graph Transformer and an implicit neural representation (INR) are employed to generate mesh-free predicted solutions. Following pretraining on data exhibiting a certain level of diversity, our model achieves zero-shot accuracies on benchmark datasets that surpass those of adequately trained expert models. Additionally, PDEformer demonstrates promising results in the inverse problem of PDE coefficient recovery.

Peer-to-peer (P2P) trading is seen as a viable solution to handle the growing number of distributed energy resources in distribution networks. However, when dealing with large-scale consumers, there are several challenges that must be addressed. One of these challenges is limited communication capabilities. Additionally, prosumers may have specific preferences when it comes to trading. Both can result in serious asynchrony in peer-to-peer trading, potentially impacting the effectiveness of negotiations and hindering convergence before the market closes. This paper introduces a connection-aware P2P trading algorithm designed for extensive prosumer trading. The algorithm facilitates asynchronous trading while respecting prosumer's autonomy in trading peer selection, an often overlooked aspect in traditional models. In addition, to optimize the use of limited connection opportunities, a smart trading peer connection selection strategy is developed to guide consumers to communicate strategically to accelerate convergence. A theoretical convergence guarantee is provided for the connection-aware P2P trading algorithm, which further details how smart selection strategies enhance convergence efficiency. Numerical studies are carried out to validate the effectiveness of the connection-aware algorithm and the performance of smart selection strategies in reducing the overall convergence time.

Widely adopted motion forecasting datasets substitute the observed sensory inputs with higher-level abstractions such as 3D boxes and polylines. These sparse shapes are inferred through annotating the original scenes with perception systems' predictions. Such intermediate representations tie the quality of the motion forecasting models to the performance of computer vision models. Moreover, the human-designed explicit interfaces between perception and motion forecasting typically pass only a subset of the semantic information present in the original sensory input. To study the effect of these modular approaches, design new paradigms that mitigate these limitations, and accelerate the development of end-to-end motion forecasting models, we augment the Waymo Open Motion Dataset (WOMD) with large-scale, high-quality, diverse LiDAR data for the motion forecasting task. The new augmented dataset WOMD-LiDAR consists of over 100,000 scenes that each spans 20 seconds, consisting of well-synchronized and calibrated high quality LiDAR point clouds captured across a range of urban and suburban geographies (//waymo.com/open/data/motion/). Compared to Waymo Open Dataset (WOD), WOMD-LiDAR dataset contains 100x more scenes. Furthermore, we integrate the LiDAR data into the motion forecasting model training and provide a strong baseline. Experiments show that the LiDAR data brings improvement in the motion forecasting task. We hope that WOMD-LiDAR will provide new opportunities for boosting end-to-end motion forecasting models.

The burgeoning field of on-device AI communication, where devices exchange information directly through embedded foundation models, such as language models (LMs), requires robust, efficient, and generalizable communication frameworks. However, integrating these frameworks with existing wireless systems and effectively managing noise and bit errors pose significant challenges. In this work, we introduce a practical on-device AI communication framework, integrated with physical layer (PHY) communication functions, demonstrated through its performance on a link-level simulator. Our framework incorporates end-to-end training with channel noise to enhance resilience, incorporates vector quantized variational autoencoders (VQ-VAE) for efficient and robust communication, and utilizes pre-trained encoder-decoder transformers for improved generalization capabilities. Simulations, across various communication scenarios, reveal that our framework achieves a 50% reduction in transmission size while demonstrating substantial generalization ability and noise robustness under standardized 3GPP channel models.

The emerging reflecting intelligent surface (RIS) technology promises to enhance the capacity of wireless communication systems via passive reflect beamforming. However, the product path loss limits its performance gains. Fully-connected (FC) active RIS, which integrates reflect-type power amplifiers into the RIS elements, has been recently introduced in response to this issue. Also, sub-connected (SC) active RIS and hybrid FC-active/passive RIS variants, which employ a limited number of reflect-type power amplifiers, have been proposed to provide energy savings. Nevertheless, their flexibility in balancing diverse capacity requirements and power consumption constraints is limited. In this direction, this study introduces novel hybrid RIS structures, wherein at least one reflecting sub-surface (RS) adopts the SC-active RIS design. The asymptotic signal-to-noise-ratio of the FC-active/passive and the proposed hybrid RIS variants is analyzed in a single-user single-input single-output setup. Furthermore, the transmit and RIS beamforming weights are jointly optimized in each scenario to maximize the energy efficiency of a hybrid RIS-aided multi-user multiple-input single-output downlink system subject to the power consumption constraints of the base station and the active RSs. Numerical simulation and analytic results highlight the performance gains of the proposed RIS designs over benchmarks, unveil non-trivial trade-offs, and provide valuable insights.

Privacy, scalability, and reliability are significant challenges in unmanned aerial vehicle (UAV) networks as distributed systems, especially when employing machine learning (ML) technologies with substantial data exchange. Recently, the application of federated learning (FL) to UAV networks has improved collaboration, privacy, resilience, and adaptability, making it a promising framework for UAV applications. However, implementing FL for UAV networks introduces drawbacks such as communication overhead, synchronization issues, scalability limitations, and resource constraints. To address these challenges, this paper presents the Blockchain-enabled Clustered and Scalable Federated Learning (BCS-FL) framework for UAV networks. This improves the decentralization, coordination, scalability, and efficiency of FL in large-scale UAV networks. The framework partitions UAV networks into separate clusters, coordinated by cluster head UAVs (CHs), to establish a connected graph. Clustering enables efficient coordination of updates to the ML model. Additionally, hybrid inter-cluster and intra-cluster model aggregation schemes generate the global model after each training round, improving collaboration and knowledge sharing among clusters. The numerical findings illustrate the achievement of convergence while also emphasizing the trade-offs between the effectiveness of training and communication efficiency.