Image token removal is an efficient augmentation strategy for reducing the cost of computing image features. However, this efficient augmentation strategy has been found to adversely affect the accuracy of CLIP-based training. We hypothesize that removing a large portion of image tokens may improperly discard the semantic content associated with a given text description, thus constituting an incorrect pairing target in CLIP training. To address this issue, we propose an attentive token removal approach for CLIP training, which retains tokens with a high semantic correlation to the text description. The correlation scores are computed in an online fashion using the EMA version of the visual encoder. Our experiments show that the proposed attentive masking approach performs better than the previous method of random token removal for CLIP training. The approach also makes it efficient to apply multiple augmentation views to the image, as well as introducing instance contrastive learning tasks between these views into the CLIP framework. Compared to other CLIP improvements that combine different pre-training targets such as SLIP and MaskCLIP, our method is not only more effective, but also much more efficient. Specifically, using ViT-B and YFCC-15M dataset, our approach achieves $43.9\%$ top-1 accuracy on ImageNet-1K zero-shot classification, as well as $62.7/42.1$ and $38.0/23.2$ I2T/T2I retrieval accuracy on Flickr30K and MS COCO, which are $+1.1\%$, $+5.5/+0.9$, and $+4.4/+1.3$ higher than the SLIP method, while being $2.30\times$ faster. An efficient version of our approach running $1.16\times$ faster than the plain CLIP model achieves significant gains of $+5.3\%$, $+11.3/+8.0$, and $+9.5/+4.9$ on these benchmarks.

In recent years, there is a surge of generation-based information extraction work, which allows a more direct use of pre-trained language models and efficiently captures output dependencies. However, previous generative methods using lexical representation do not naturally fit document-level relation extraction (DocRE) where there are multiple entities and relational facts. In this paper, we investigate the root cause of the underwhelming performance of the existing generative DocRE models and discover that the culprit is the inadequacy of the training paradigm, instead of the capacities of the models. We propose to generate a symbolic and ordered sequence from the relation matrix which is deterministic and easier for model to learn. Moreover, we design a parallel row generation method to process overlong target sequences. Besides, we introduce several negative sampling strategies to improve the performance with balanced signals. Experimental results on four datasets show that our proposed method can improve the performance of the generative DocRE models. We have released our code at https://github.com/ayyyq/DORE.

In split machine learning (ML), different partitions of a neural network (NN) are executed by different computing nodes, requiring a large amount of communication cost. To ease communication burden, over-the-air computation (OAC) can efficiently implement all or part of the computation at the same time of communication. Based on the proposed system, the system implementation over wireless network is introduced and we provide the problem formulation. In particular, we show that the inter-layer connection in a NN of any size can be mathematically decomposed into a set of linear precoding and combining transformations over MIMO channels. Therefore, the precoding matrix at the transmitter and the combining matrix at the receiver of each MIMO link, as well as the channel matrix itself, can jointly serve as a fully connected layer of the NN. The generalization of the proposed scheme to the conventional NNs is also introduced. Finally, we extend the proposed scheme to the widely used convolutional neural networks and demonstrate its effectiveness under both the static and quasi-static memory channel conditions with comprehensive simulations. In such a split ML system, the precoding and combining matrices are regarded as trainable parameters, while MIMO channel matrix is regarded as unknown (implicit) parameters.

近年来，基于深度学习的模型在视频超分辨率（VSR）方面取得了显着性能，但是这些模型中的大多数不适用于在线视频应用程序。这些方法仅考虑失真质量，而忽略了在线应用程序的关键要求，例如低延迟和模型较低的复杂性。在本文中，我们专注于在线视频传输，其中需要VSR算法来实时生成高分辨率的视频序列。为了应对此类挑战，我们提出了一种基于一种新的内核知识转移方法，称为卷积核旁路移植物（CKBG）。首先，我们设计了一个轻巧的网络结构，该结构不需要将来的帧作为输入，并节省了缓存这些帧的额外时间成本。然后，我们提出的CKBG方法通过用``核移植物）''绕过原始网络来增强这种轻巧的基础模型，这些网络是包含外部预验证图像SR模型的先验知识的额外卷积内核。在测试阶段，我们通过将其转换为简单的单路结构来进一步加速移植的多支球网络。实验结果表明，我们提出的方法可以处理高达110 fps的在线视频序列，并且模型复杂性非常低和竞争性SR性能。

时间序列数据生成近年来越来越受到关注。已经提出了几种生成的对抗网络（GaN）的方法通常是假设目标时间序列数据良好格式化并完成的假设来解决问题。然而，现实世界时间序列（RTS）数据远离该乌托邦，例如，具有可变长度的长序列和信息缺失数据，用于设计强大的发电算法的棘手挑战。在本文中，我们向RTS数据提出了一种新的生成框架 - RTSGAN来解决上述挑战。 RTSGAN首先学习编码器 - 解码器模块，该模块提供时间序列实例和固定维度潜在载体之间的映射，然后学习生成模块以在同一潜在空间中生成vectors。通过组合发电机和解码器，RTSGAN能够生成尊重原始特征分布和时间动态的RTS。为了生成具有缺失值的时间序列，我们进一步用观察嵌入层和决定和生成解码器装备了RTSGAN，以更好地利用信息缺失模式。四个RTS数据集上的实验表明，该框架在用于下游分类和预测任务的合成数据实用程序方面优于前一代方法。

在移动边缘网络上部署深神经网络（DNN）的主要挑战是如何分离DNN模型，以匹配网络架构以及所有节点的计算和通信容量。这基本上涉及两个高耦合程序：模型生成和模型分裂。在本文中，提出了一种联合模型分割和神经结构搜索（JMSNAS）框架以在移动边缘网络上自动生成和部署DNN模型。考虑到计算和通信资源约束，配制计算图形搜索问题以查找DNN模型的多分裂点，然后培训模型以满足一些精度要求。此外，通过正确设计目标函数来实现模型精度和完成延迟之间的权衡。实验结果证实了通过最先进的分机学习设计方法的提出框架的优越性。

神经体系结构搜索（NAS）的主要挑战之一是有效地对体系结构的性能进行排名。绩效排名者的主流评估使用排名相关性（例如，肯德尔的tau），这对整个空间都同样关注。但是，NAS的优化目标是识别顶级体系结构，同时对搜索空间中其他体系结构的关注更少。在本文中，我们从经验和理论上都表明，标准化的累积累积增益（NDCG）对于排名者来说是一个更好的指标。随后，我们提出了一种新算法Acenas，该算法直接通过Lambdarank优化NDCG。它还利用体重共享NAS产生的弱标签来预先培训排名，以便进一步降低搜索成本。对12个NAS基准和大规模搜索空间进行的广泛实验表明，我们的方法始终超过SOTA NAS方法，精度提高了3.67％，搜索成本降低了8倍。

Recent years have witnessed the rapid growth of Small Private Online Courses (SPOC) which is able to highly customized and personalized to adapt variable educational requests, in which machine learning techniques are explored to summarize and predict the learner's performance, mostly focus on the final grade. However, the problem is that the final grade of learners on SPOC is generally seriously imbalance which handicaps the training of prediction model. To solve this problem, a sampling batch normalization embedded deep neural network (SBNEDNN) method is developed in this paper. First, a combined indicator is defined to measure the distribution of the data, then a rule is established to guide the sampling process. Second, the batch normalization (BN) modified layers are embedded into full connected neural network to solve the data imbalanced problem. Experimental results with other three deep learning methods demonstrates the superiority of the proposed method.

The ''Propose-Test-Release'' (PTR) framework is a classic recipe for designing differentially private (DP) algorithms that are data-adaptive, i.e. those that add less noise when the input dataset is nice. We extend PTR to a more general setting by privately testing data-dependent privacy losses rather than local sensitivity, hence making it applicable beyond the standard noise-adding mechanisms, e.g. to queries with unbounded or undefined sensitivity. We demonstrate the versatility of generalized PTR using private linear regression as a case study. Additionally, we apply our algorithm to solve an open problem from ''Private Aggregation of Teacher Ensembles (PATE)'' -- privately releasing the entire model with a delicate data-dependent analysis.

Full-body reconstruction is a fundamental but challenging task. Owing to the lack of annotated data, the performances of existing methods are largely limited. In this paper, we propose a novel method named Full-body Reconstruction from Part Experts~(FuRPE) to tackle this issue. In FuRPE, the network is trained using pseudo labels and features generated from part-experts. An simple yet effective pseudo ground-truth selection scheme is proposed to extract high-quality pseudo labels. In this way, a large-scale of existing human body reconstruction datasets can be leveraged and contribute to the model training. In addition, an exponential moving average training strategy is introduced to train the network in a self-supervised manner, further boosting the performance of the model. Extensive experiments on several widely used datasets demonstrate the effectiveness of our method over the baseline. Our method achieves the state-of-the-art performance. Code will be publicly available for further research.