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

The execution of large deep neural networks (DNN) at mobile edge devices requires considerable consumption of critical resources, such as energy, while imposing demands on hardware capabilities. In approaches based on edge computing the execution of the models is offloaded to a compute-capable device positioned at the edge of 5G infrastructures. The main issue of the latter class of approaches is the need to transport information-rich signals over wireless links with limited and time-varying capacity. The recent split computing paradigm attempts to resolve this impasse by distributing the execution of DNN models across the layers of the systems to reduce the amount of data to be transmitted while imposing minimal computing load on mobile devices. In this context, we propose a novel split computing approach based on slimmable ensemble encoders. The key advantage of our design is the ability to adapt computational load and transmitted data size in real-time with minimal overhead and time. This is in contrast with existing approaches, where the same adaptation requires costly context switching and model loading. Moreover, our model outperforms existing solutions in terms of compression efficacy and execution time, especially in the context of weak mobile devices. We present a comprehensive comparison with the most advanced split computing solutions, as well as an experimental evaluation on GPU-less devices.

Federated learning (FL) is a common and practical framework for learning a machine model in a decentralized fashion. A primary motivation behind this decentralized approach is data privacy, ensuring that the learner never sees the data of each local source itself. Federated learning then comes with two majors challenges: one is handling potentially complex model updates between a server and a large number of data sources; the other is that de-centralization may, in fact, be insufficient for privacy, as the local updates themselves can reveal information about the sources' data. To address these issues, we consider an approach to federated learning that combines quantization and differential privacy. Absent privacy, Federated Learning often relies on quantization to reduce communication complexity. We build upon this approach and develop a new algorithm called the \textbf{R}andomized \textbf{Q}uantization \textbf{M}echanism (RQM), which obtains privacy through a two-levels of randomization. More precisely, we randomly sub-sample feasible quantization levels, then employ a randomized rounding procedure using these sub-sampled discrete levels. We are able to establish that our results preserve ``Renyi differential privacy'' (Renyi DP). We empirically study the performance of our algorithm and demonstrate that compared to previous work it yields improved privacy-accuracy trade-offs for DP federated learning. To the best of our knowledge, this is the first study that solely relies on randomized quantization without incorporating explicit discrete noise to achieve Renyi DP guarantees in Federated Learning systems.

A sum-factorization form for the evaluation of Hadamard products with a tensor product basis is derived in this work. The proposed algorithm allows for Hadamard products to be computed in $\mathcal{O}\left(n^{d+1}\right)$ flops rather than $\mathcal{O}\left(n^{2d}\right)$, where $d$ is the dimension of the problem. With this improvement, entropy conserving and stable schemes, that require a dense Hadamard product in the general modal case, become computationally competitive with the modal discontinuous Galerkin (DG) scheme. We numerically demonstrate the application of the sum-factorized Hadamard product in our in-house partial differential equation solver PHiLiP based on the Nonlinearly Stable Flux Reconstruction scheme. We demonstrate that the entropy conserving flow solver scales at $\mathcal{O}\left(n^{d+1}\right)$ for three-dimensional compressible flow in curvilinear coordinates, along with a computational cost comparison with the modal DG and over-integrated DG schemes.

A growing line of work shows how learned predictions can be used to break through worst-case barriers to improve the running time of an algorithm. However, incorporating predictions into data structures with strong theoretical guarantees remains underdeveloped. This paper takes a step in this direction by showing that predictions can be leveraged in the fundamental online list labeling problem. In the problem, n items arrive over time and must be stored in sorted order in an array of size Theta(n). The array slot of an element is its label and the goal is to maintain sorted order while minimizing the total number of elements moved (i.e., relabeled). We design a new list labeling data structure and bound its performance in two models. In the worst-case learning-augmented model, we give guarantees in terms of the error in the predictions. Our data structure provides strong guarantees: it is optimal for any prediction error and guarantees the best-known worst-case bound even when the predictions are entirely erroneous. We also consider a stochastic error model and bound the performance in terms of the expectation and variance of the error. Finally, the theoretical results are demonstrated empirically. In particular, we show that our data structure has strong performance on real temporal data sets where predictions are constructed from elements that arrived in the past, as is typically done in a practical use case.

Active muscles are crucial for maintaining postural stability when seated in a moving vehicle. Advanced active 3D non-linear full body models have been developed for impact and comfort simulation, including large numbers of individual muscle elements, and detailed non-linear models of the joint structures. While such models have an apparent potential to provide insight into postural stabilization, they are computationally demanding, making them less practical in particular for driving comfort where long time periods are to be studied. In vibration comfort and in general biomechanical research, linearized models are effectively used. This paper evaluates the effectiveness of simplified 3D full body human models to capture vibration comfort. An efficient seated human body model is developed and validated using experimental data. We evaluate the required complexity in terms of joints and degrees of freedom for the spine, and explore how well linear spring-damper models can approximate reflexive postural stabilization. Results indicate that linear stiffness and damping models can capture well the human response. However, the results are improved by adding proportional integral derivative (PID) controllers to maintain the defined initial body posture. The integrator is shown to be essential to prevent drift from the defined posture. The joint angular relative displacement is used as the input reference to each PID controller. With this model, a faster than real time solution is obtained when used with a simple seat model. The paper also discusses advantages and disadvantages of various models and provides insight into which models are more appropriate for motion comfort analysis. The results of this paper can provide valuable insights for designers, and researchers in the automotive and seating industries to improve the comfort and safety of seats and vehicles occupants.

Context: Due to the association of significant efforts, even a minor improvement in the effectiveness of Code Reviews(CR) can incur significant savings for a software development organization. Aim: This study aims to develop a finer grain understanding of what makes a code review comment useful to OSS developers, to what extent a code review comment is considered useful to them, and how various contextual and participant-related factors influence its usefulness level. Method: On this goal, we have conducted a three-stage mixed-method study. We randomly selected 2,500 CR comments from the OpenDev Nova project and manually categorized the comments. We designed a survey of OpenDev developers to better understand their perspectives on useful CRs. Combining our survey-obtained scores with our manually labeled dataset, we trained two regression models - one to identify factors that influence the usefulness of CR comments and the other to identify factors that improve the odds of `Functional' defect identification over the others. Key findings: The results of our study suggest that a CR comment's usefulness is dictated not only by its technical contributions such as defect findings or quality improvement tips but also by its linguistic characteristics such as comprehensibility and politeness. While a reviewer's coding experience positively associates with CR usefulness, the number of mutual reviews, comment volume in a file, the total number of lines added /modified, and CR interval has the opposite associations. While authorship and reviewership experiences for the files under review have been the most popular attributes for reviewer recommendation systems, we do not find any significant association of those attributes with CR usefulness.

Mutual information is a general statistical dependency measure which has found applications in representation learning, causality, domain generalization and computational biology. However, mutual information estimators are typically evaluated on simple families of probability distributions, namely multivariate normal distribution and selected distributions with one-dimensional random variables. In this paper, we show how to construct a diverse family of distributions with known ground-truth mutual information and propose a language-independent benchmarking platform for mutual information estimators. We discuss the general applicability and limitations of classical and neural estimators in settings involving high dimensions, sparse interactions, long-tailed distributions, and high mutual information. Finally, we provide guidelines for practitioners on how to select appropriate estimator adapted to the difficulty of problem considered and issues one needs to consider when applying an estimator to a new data set.

Emerging connected vehicle (CV) data sets have recently become commercially available. This paper presents several tools using CV data to evaluate traffic progression quality along a signalized corridor. These include both performance measures for high-level analysis as well as visualizations to examine details of the coordinated operation. With the use of CV data, it is possible to assess not only the movement of traffic on the corridor but also to consider its origin-destination (OD) path through the corridor. Results for the real-world operation of an eight-intersection signalized arterial are presented. A series of high-level performance measures are used to evaluate overall performance by time of day, with differing results by metric. Next, the details of the operation are examined with the use of two visualization tools: a cyclic time-space diagram (TSD) and an empirical platoon progression diagram (PPD). Comparing flow visualizations developed with different included OD paths reveals several features. In addition, speed heat maps are generated, providing both speed performance along the corridor. The proposed visualization tools portray the corridor performance holistically instead of combining individual signal performance metrics. The techniques exhibited in this study are compelling for identifying locations where engineering solutions are required. The recent progress in infrastructure-free sensing technology has significantly increased the scope of CV data-based traffic management systems. The study demonstrates the utility of CV trajectory data for obtaining high-level details of the corridor performance and drilling down into the minute specifics.

Communication plays a vital role in multi-agent systems, fostering collaboration and coordination. However, in real-world scenarios where communication is bandwidth-limited, existing multi-agent reinforcement learning (MARL) algorithms often provide agents with a binary choice: either transmitting a fixed number of bytes or no information at all. This limitation hinders the ability to effectively utilize the available bandwidth. To overcome this challenge, we present the Dynamic Size Message Scheduling (DSMS) method, which introduces a finer-grained approach to scheduling by considering the actual size of the information to be exchanged. Our contribution lies in adaptively adjusting message sizes using Fourier transform-based compression techniques, enabling agents to tailor their messages to match the allocated bandwidth while striking a balance between information loss and transmission efficiency. Receiving agents can reliably decompress the messages using the inverse Fourier transform. Experimental results demonstrate that DSMS significantly improves performance in multi-agent cooperative tasks by optimizing the utilization of bandwidth and effectively balancing information value.

We propose a family of recursive cutting-plane algorithms to solve feasibility problems with constrained memory, which can also be used for first-order convex optimization. Precisely, in order to find a point within a ball of radius $\epsilon$ with a separation oracle in dimension $d$ -- or to minimize $1$-Lipschitz convex functions to accuracy $\epsilon$ over the unit ball -- our algorithms use $\mathcal O(\frac{d^2}{p}\ln \frac{1}{\epsilon})$ bits of memory, and make $\mathcal O((C\frac{d}{p}\ln \frac{1}{\epsilon})^p)$ oracle calls, for some universal constant $C \geq 1$. The family is parametrized by $p\in[d]$ and provides an oracle-complexity/memory trade-off in the sub-polynomial regime $\ln\frac{1}{\epsilon}\gg\ln d$. While several works gave lower-bound trade-offs (impossibility results) -- we explicit here their dependence with $\ln\frac{1}{\epsilon}$, showing that these also hold in any sub-polynomial regime -- to the best of our knowledge this is the first class of algorithms that provides a positive trade-off between gradient descent and cutting-plane methods in any regime with $\epsilon\leq 1/\sqrt d$. The algorithms divide the $d$ variables into $p$ blocks and optimize over blocks sequentially, with approximate separation vectors constructed using a variant of Vaidya's method. In the regime $\epsilon \leq d^{-\Omega(d)}$, our algorithm with $p=d$ achieves the information-theoretic optimal memory usage and improves the oracle-complexity of gradient descent.