无人驾驶飞机（UAV）的实时对象检测是一个具有挑战性的问题，因为Edge GPU设备作为物联网（IoT）节点的计算资源有限。为了解决这个问题，在本文中，我们提出了一种基于Yolox模型的新型轻型深度学习体系结构，用于Edge GPU上的实时对象检测。首先，我们设计了一个有效且轻巧的PixSF头，以更换Yolox的原始头部以更好地检测小物体，可以将其进一步嵌入深度可分离的卷积（DS Conv）中，以达到更轻的头。然后，开发为减少网络参数的颈层中的较小结构，这是精度和速度之间的权衡。此外，我们将注意模块嵌入头层中，以改善预测头的特征提取效果。同时，我们还改进了标签分配策略和损失功能，以减轻UAV数据集的类别不平衡和盒子优化问题。最后，提出了辅助头进行在线蒸馏，以提高PIXSF Head中嵌入位置嵌入和特征提取的能力。在NVIDIA Jetson NX和Jetson Nano GPU嵌入平台上，我们的轻质模型的性能得到了实验验证。扩展的实验表明，与目前的模型相比，Fasterx模型在Visdrone2021数据集中实现了更好的折衷和延迟之间的折衷。

对看不见的环境变化的深入强化学习的概括通常需要对大量各种培训变化进行政策学习。我们从经验上观察到，接受过许多变化的代理商（通才）倾向于在一开始就更快地学习，但是长期以来其最佳水平的性能高原。相比之下，只接受一些变体培训的代理商（专家）通常可以在有限的计算预算下获得高回报。为了两全其美，我们提出了一个新颖的通才特权训练框架。具体来说，我们首先培训一名通才的所有环境变化。当它无法改善时，我们会推出大量的专家，并从通才克隆过重量，每个人都接受了训练，以掌握选定的一小部分变化子集。我们终于通过所有专家的示范引起的辅助奖励恢复了通才的培训。特别是，我们调查了开始专业培训的时机，并在专家的帮助下比较策略以学习通才。我们表明，该框架将政策学习的信封推向了包括Procgen，Meta-World和Maniskill在内的几个具有挑战性和流行的基准。

3D视觉输入的对象操纵对构建可宽大的感知和政策模型构成了许多挑战。然而，现有基准中的3D资产主要缺乏与拓扑和几何中的现实世界内复杂的3D形状的多样性。在这里，我们提出了Sapien操纵技能基准（Manishill）以在全物理模拟器中的各种物体上基准操纵技巧。 Manishill中的3D资产包括大型课堂内拓扑和几何变化。仔细选择任务以涵盖不同类型的操纵挑战。 3D Vision的最新进展也使我们认为我们应该定制基准，以便挑战旨在邀请研究3D深入学习的研究人员。为此，我们模拟了一个移动的全景摄像头，返回以自我为中心的点云或RGB-D图像。此外，我们希望Manishill是为一个对操纵研究感兴趣的广泛研究人员提供服务。除了支持从互动的政策学习，我们还支持学习 - 从演示（LFD）方法，通过提供大量的高质量演示（〜36,000个成功的轨迹，总共〜1.5米点云/ RGB-D帧）。我们提供使用3D深度学习和LFD算法的基线。我们的基准（模拟器，环境，SDK和基线）的所有代码都是开放的，并且将基于基准举办跨学科研究人员面临的挑战。

由于其令人鼓舞的性能，在各种控制任务中的令人鼓舞的表现，深增强学习（Deep RL）一直在受到更高的关注。然而，在训练神经网络中的常规正则化技术（例如，$ L_2 $正则化，辍学）已经在RL方法中被忽略，可能是因为代理通常在相同的环境中进行培训和评估，因为Deep RL社区重点关注更多-Level算法设计。在这项工作中，我们在连续控制任务中提出了具有多种策略优化算法的正则化技术的第一综合研究。有趣的是，我们发现策略网络上的传统正则化技术通常可以带来大量改进，特别是在更难的任务上。我们的研究结果显示在训练HyperParameter变化方面是强大的。我们还将这些技术与更广泛使用的熵正则化进行了比较。此外，我们还研究正规化不同的组件，并发现策略网络通常是最佳的。我们进一步分析了为什么正则化可能有助于从四个观点来帮助推广 - 样本复杂性，奖励分配，重量规范和噪音鲁棒性。我们希望我们的研究为未来的规则策略优化算法提供指导。我们的代码可在https://github.com/xuanlinli17/ICLRR2021_RLREG上获得。

The growing interest in intelligent services and privacy protection for mobile devices has given rise to the widespread application of federated learning in Multi-access Edge Computing (MEC). Diverse user behaviors call for personalized services with heterogeneous Machine Learning (ML) models on different devices. Federated Multi-task Learning (FMTL) is proposed to train related but personalized ML models for different devices, whereas previous works suffer from excessive communication overhead during training and neglect the model heterogeneity among devices in MEC. Introducing knowledge distillation into FMTL can simultaneously enable efficient communication and model heterogeneity among clients, whereas existing methods rely on a public dataset, which is impractical in reality. To tackle this dilemma, Federated MultI-task Distillation for Multi-access Edge CompuTing (FedICT) is proposed. FedICT direct local-global knowledge aloof during bi-directional distillation processes between clients and the server, aiming to enable multi-task clients while alleviating client drift derived from divergent optimization directions of client-side local models. Specifically, FedICT includes Federated Prior Knowledge Distillation (FPKD) and Local Knowledge Adjustment (LKA). FPKD is proposed to reinforce the clients' fitting of local data by introducing prior knowledge of local data distributions. Moreover, LKA is proposed to correct the distillation loss of the server, making the transferred local knowledge better match the generalized representation. Experiments on three datasets show that FedICT significantly outperforms all compared benchmarks in various data heterogeneous and model architecture settings, achieving improved accuracy with less than 1.2% training communication overhead compared with FedAvg and no more than 75% training communication round compared with FedGKT.

In this paper we study the smooth strongly convex minimization problem $\min_{x}\min_y f(x,y)$. The existing optimal first-order methods require $\mathcal{O}(\sqrt{\max\{\kappa_x,\kappa_y\}} \log 1/\epsilon)$ of computations of both $\nabla_x f(x,y)$ and $\nabla_y f(x,y)$, where $\kappa_x$ and $\kappa_y$ are condition numbers with respect to variable blocks $x$ and $y$. We propose a new algorithm that only requires $\mathcal{O}(\sqrt{\kappa_x} \log 1/\epsilon)$ of computations of $\nabla_x f(x,y)$ and $\mathcal{O}(\sqrt{\kappa_y} \log 1/\epsilon)$ computations of $\nabla_y f(x,y)$. In some applications $\kappa_x \gg \kappa_y$, and computation of $\nabla_y f(x,y)$ is significantly cheaper than computation of $\nabla_x f(x,y)$. In this case, our algorithm substantially outperforms the existing state-of-the-art methods.

Feature transformation for AI is an essential task to boost the effectiveness and interpretability of machine learning (ML). Feature transformation aims to transform original data to identify an optimal feature space that enhances the performances of a downstream ML model. Existing studies either combines preprocessing, feature selection, and generation skills to empirically transform data, or automate feature transformation by machine intelligence, such as reinforcement learning. However, existing studies suffer from: 1) high-dimensional non-discriminative feature space; 2) inability to represent complex situational states; 3) inefficiency in integrating local and global feature information. To fill the research gap, we formulate the feature transformation task as an iterative, nested process of feature generation and selection, where feature generation is to generate and add new features based on original features, and feature selection is to remove redundant features to control the size of feature space. Finally, we present extensive experiments and case studies to illustrate 24.7\% improvements in F1 scores compared with SOTAs and robustness in high-dimensional data.

In recent years, large amounts of effort have been put into pushing forward the real-world application of dynamic digital human (DDH). However, most current quality assessment research focuses on evaluating static 3D models and usually ignores motion distortions. Therefore, in this paper, we construct a large-scale dynamic digital human quality assessment (DDH-QA) database with diverse motion content as well as multiple distortions to comprehensively study the perceptual quality of DDHs. Both model-based distortion (noise, compression) and motion-based distortion (binding error, motion unnaturalness) are taken into consideration. Ten types of common motion are employed to drive the DDHs and a total of 800 DDHs are generated in the end. Afterward, we render the video sequences of the distorted DDHs as the evaluation media and carry out a well-controlled subjective experiment. Then a benchmark experiment is conducted with the state-of-the-art video quality assessment (VQA) methods and the experimental results show that existing VQA methods are limited in assessing the perceptual loss of DDHs. The database will be made publicly available to facilitate future research.

Recently, over-height vehicle strike frequently occurs, causing great economic cost and serious safety problems. Hence, an alert system which can accurately discover any possible height limiting devices in advance is necessary to be employed in modern large or medium sized cars, such as touring cars. Detecting and estimating the height limiting devices act as the key point of a successful height limit alert system. Though there are some works research height limit estimation, existing methods are either too computational expensive or not accurate enough. In this paper, we propose a novel stereo-based pipeline named SHLE for height limit estimation. Our SHLE pipeline consists of two stages. In stage 1, a novel devices detection and tracking scheme is introduced, which accurately locate the height limit devices in the left or right image. Then, in stage 2, the depth is temporally measured, extracted and filtered to calculate the height limit device. To benchmark the height limit estimation task, we build a large-scale dataset named "Disparity Height", where stereo images, pre-computed disparities and ground-truth height limit annotations are provided. We conducted extensive experiments on "Disparity Height" and the results show that SHLE achieves an average error below than 10cm though the car is 70m away from the devices. Our method also outperforms all compared baselines and achieves state-of-the-art performance. Code is available at https://github.com/Yang-Kaixing/SHLE.

Steering language generation towards objectives or away from undesired content has been a long-standing goal in utilizing language models (LM). Recent work has demonstrated reinforcement learning and weighted decoding as effective approaches to achieve a higher level of language control and quality with pros and cons. In this work, we propose a novel critic decoding method for controlled language generation (CriticControl) that combines the strengths of reinforcement learning and weighted decoding. Specifically, we adopt the actor-critic framework to train an LM-steering critic from non-differentiable reward models. And similar to weighted decoding, our method freezes the language model and manipulates the output token distribution using called critic, improving training efficiency and stability. Evaluation of our method on three controlled generation tasks, namely topic control, sentiment control, and detoxification, shows that our approach generates more coherent and well-controlled texts than previous methods. In addition, CriticControl demonstrates superior generalization ability in zero-shot settings. Human evaluation studies also corroborate our findings.