To address privacy concerns and reduce network latency, there has been a recent trend of compressing cumbersome recommendation models trained on the cloud and deploying compact recommender models to resource-limited devices for real-time recommendation. Existing solutions generally overlook device heterogeneity and user heterogeneity. They either require all devices to share the same compressed model or the devices with the same resource budget to share the same model. However, even users with the same devices may have different preferences. In addition, they assume the available resources (e.g., memory) for the recommender on a device are constant, which is not reflective of reality. In light of device and user heterogeneities as well as dynamic resource constraints, this paper proposes a Personalized Elastic Embedding Learning framework (PEEL) for on-device recommendation, which generates personalized embeddings for devices with various memory budgets in once-for-all manner, efficiently adapting to new or dynamic budgets, and effectively addressing user preference diversity by assigning personalized embeddings for different groups of users. Specifically, it pretrains using user-item interaction instances to generate the global embedding table and cluster users into groups. Then, it refines the embedding tables with local interaction instances within each group. Personalized elastic embedding is generated from the group-wise embedding blocks and their weights that indicate the contribution of each embedding block to the local recommendation performance. PEEL efficiently generates personalized elastic embeddings by selecting embedding blocks with the largest weights, making it adaptable to dynamic memory budgets. Extensive experiments are conducted on two public datasets, and the results show that PEEL yields superior performance on devices with heterogeneous and dynamic memory budgets.
相關內容
IFIP TC13 Conference on Human-Computer Interaction是人機交互領域的研究者和實踐者展示其工作的重要平臺。多年來，這些會議吸引了來自幾個國家和文化的研究人員。官網鏈接： ·
神經元
·
剪枝
·
MoDELS
·
Processing（編程語言）
·
Deep neural networks (DNNs) have been found to be vulnerable to backdoor attacks, raising security concerns about their deployment in mission-critical applications. While existing defense methods have demonstrated promising results, it is still not clear how to effectively remove backdoor-associated neurons in backdoored DNNs. In this paper, we propose a novel defense called \emph{Reconstructive Neuron Pruning} (RNP) to expose and prune backdoor neurons via an unlearning and then recovering process. Specifically, RNP first unlearns the neurons by maximizing the model's error on a small subset of clean samples and then recovers the neurons by minimizing the model's error on the same data. In RNP, unlearning is operated at the neuron level while recovering is operated at the filter level, forming an asymmetric reconstructive learning procedure. We show that such an asymmetric process on only a few clean samples can effectively expose and prune the backdoor neurons implanted by a wide range of attacks, achieving a new state-of-the-art defense performance. Moreover, the unlearned model at the intermediate step of our RNP can be directly used to improve other backdoor defense tasks including backdoor removal, trigger recovery, backdoor label detection, and backdoor sample detection. Code is available at \url{//github.com/bboylyg/RNP}.
Uniformly valid inference for cointegrated vector autoregressive processes has so far proven difficult due to certain discontinuities arising in the asymptotic distribution of the least squares estimator. We extend asymptotic results from the univariate case to multiple dimensions and show how inference can be based on these results. Furthermore, we show that lag augmentation and a recent instrumental variable procedure can also yield uniformly valid tests and confidence regions. We verify the theoretical findings and investigate finite sample properties in simulation experiments for two specific examples.
Modular exponentiation and scalar multiplication are important operations of most public key cryptosystems, and their fast calculation is essential to improve the system efficiency. The shortest addition chain is one of the most important mathematical concepts to realize the optimization. However, finding a shortest addition chain of length k is an NP-hard problem, whose time complexity is comparable to O($k!$). This paper proposes some novel methods to generate short addition chains. We firstly present a Simplified Power-tree method by deeply deleting the power-tree, whose time complexity is reduced to O($k^2$) sacrificing some increasing of the addition chain length. Moreover, a Cross Window method and its variant are introduced by improving the Window method. More precisely, the Cross Window method uses the cross correlation to deal with the windows and its pre-computation is optimized by the Addition Sequence algorithm. The theoretical analysis is conducted to show the correctness and effectiveness. Meanwhile, the experiment shows that the new methods can obtain shorter addition chains compared to the existing methods. The Cross Window method with the Addition Sequence algorithm can attain 9.5% reduction of the addition chain length, in the best case, compared to the Window method.
In a recent work, Chen, Hoza, Lyu, Tal and Wu (FOCS 2023) showed an improved error reduction framework for the derandomization of regular read-once branching programs (ROBPs). Their result is based on a clever modification to the inverse Laplacian perspective of space-bounded derandomization, which was originally introduced by Ahmadinejad, Kelner, Murtagh, Peebles, Sidford and Vadhan (FOCS 2020). In this work, we give an alternative error reduction framework for regular ROBPs. Our new framework is based on a binary recursive formula from the work of Chattopadhyay and Liao (CCC 2020), that they used to construct weighted pseudorandom generators (WPRGs) for general ROBPs. Based on our new error reduction framework, we give alternative proofs to the following results for regular ROBPs of length $n$ and width $w$, both of which were proved in the work of Chen et al. using their error reduction: $\bullet$ There is a WPRG with error $\varepsilon$ that has seed length $\tilde{O}(\log(n)(\sqrt{\log(1/\varepsilon)}+\log(w))+\log(1/\varepsilon)).$ $\bullet$ There is a (non-black-box) deterministic algorithm which estimates the expectation of any such program within error $\pm\varepsilon$ with space complexity $\tilde{O}(\log(nw)\cdot\log\log(1/\varepsilon)).$ (This was first proved in the work of Ahmadinejad et al., but the proof by Chen et al. is simpler.) Because of the binary recursive nature of our new framework, both of our proofs are based on a straightforward induction that is arguably simpler than the Laplacian-based proof in the work of Chen et al.
In semi-supervised domain adaptation, a few labeled samples per class in the target domain guide features of the remaining target samples to aggregate around them. However, the trained model cannot produce a highly discriminative feature representation for the target domain because the training data is dominated by labeled samples from the source domain. This could lead to disconnection between the labeled and unlabeled target samples as well as misalignment between unlabeled target samples and the source domain. In this paper, we propose a novel approach called Cross-domain Adaptive Clustering to address this problem. To achieve both inter-domain and intra-domain adaptation, we first introduce an adversarial adaptive clustering loss to group features of unlabeled target data into clusters and perform cluster-wise feature alignment across the source and target domains. We further apply pseudo labeling to unlabeled samples in the target domain and retain pseudo-labels with high confidence. Pseudo labeling expands the number of ``labeled" samples in each class in the target domain, and thus produces a more robust and powerful cluster core for each class to facilitate adversarial learning. Extensive experiments on benchmark datasets, including DomainNet, Office-Home and Office, demonstrate that our proposed approach achieves the state-of-the-art performance in semi-supervised domain adaptation.
Leveraging datasets available to learn a model with high generalization ability to unseen domains is important for computer vision, especially when the unseen domain's annotated data are unavailable. We study a novel and practical problem of Open Domain Generalization (OpenDG), which learns from different source domains to achieve high performance on an unknown target domain, where the distributions and label sets of each individual source domain and the target domain can be different. The problem can be generally applied to diverse source domains and widely applicable to real-world applications. We propose a Domain-Augmented Meta-Learning framework to learn open-domain generalizable representations. We augment domains on both feature-level by a new Dirichlet mixup and label-level by distilled soft-labeling, which complements each domain with missing classes and other domain knowledge. We conduct meta-learning over domains by designing new meta-learning tasks and losses to preserve domain unique knowledge and generalize knowledge across domains simultaneously. Experiment results on various multi-domain datasets demonstrate that the proposed Domain-Augmented Meta-Learning (DAML) outperforms prior methods for unseen domain recognition.
Social relations are often used to improve recommendation quality when user-item interaction data is sparse in recommender systems. Most existing social recommendation models exploit pairwise relations to mine potential user preferences. However, real-life interactions among users are very complicated and user relations can be high-order. Hypergraph provides a natural way to model complex high-order relations, while its potentials for improving social recommendation are under-explored. In this paper, we fill this gap and propose a multi-channel hypergraph convolutional network to enhance social recommendation by leveraging high-order user relations. Technically, each channel in the network encodes a hypergraph that depicts a common high-order user relation pattern via hypergraph convolution. By aggregating the embeddings learned through multiple channels, we obtain comprehensive user representations to generate recommendation results. However, the aggregation operation might also obscure the inherent characteristics of different types of high-order connectivity information. To compensate for the aggregating loss, we innovatively integrate self-supervised learning into the training of the hypergraph convolutional network to regain the connectivity information with hierarchical mutual information maximization. The experimental results on multiple real-world datasets show that the proposed model outperforms the SOTA methods, and the ablation study verifies the effectiveness of the multi-channel setting and the self-supervised task. The implementation of our model is available via //github.com/Coder-Yu/RecQ.
Learning latent representations of nodes in graphs is an important and ubiquitous task with widespread applications such as link prediction, node classification, and graph visualization. Previous methods on graph representation learning mainly focus on static graphs, however, many real-world graphs are dynamic and evolve over time. In this paper, we present Dynamic Self-Attention Network (DySAT), a novel neural architecture that operates on dynamic graphs and learns node representations that capture both structural properties and temporal evolutionary patterns. Specifically, DySAT computes node representations by jointly employing self-attention layers along two dimensions: structural neighborhood and temporal dynamics. We conduct link prediction experiments on two classes of graphs: communication networks and bipartite rating networks. Our experimental results show that DySAT has a significant performance gain over several different state-of-the-art graph embedding baselines.
Benefit from the quick development of deep learning techniques, salient object detection has achieved remarkable progresses recently. However, there still exists following two major challenges that hinder its application in embedded devices, low resolution output and heavy model weight. To this end, this paper presents an accurate yet compact deep network for efficient salient object detection. More specifically, given a coarse saliency prediction in the deepest layer, we first employ residual learning to learn side-output residual features for saliency refinement, which can be achieved with very limited convolutional parameters while keep accuracy. Secondly, we further propose reverse attention to guide such side-output residual learning in a top-down manner. By erasing the current predicted salient regions from side-output features, the network can eventually explore the missing object parts and details which results in high resolution and accuracy. Experiments on six benchmark datasets demonstrate that the proposed approach compares favorably against state-of-the-art methods, and with advantages in terms of simplicity, efficiency (45 FPS) and model size (81 MB).
Collaborative filtering often suffers from sparsity and cold start problems in real recommendation scenarios, therefore, researchers and engineers usually use side information to address the issues and improve the performance of recommender systems. In this paper, we consider knowledge graphs as the source of side information. We propose MKR, a Multi-task feature learning approach for Knowledge graph enhanced Recommendation. MKR is a deep end-to-end framework that utilizes knowledge graph embedding task to assist recommendation task. The two tasks are associated by cross&compress units, which automatically share latent features and learn high-order interactions between items in recommender systems and entities in the knowledge graph. We prove that cross&compress units have sufficient capability of polynomial approximation, and show that MKR is a generalized framework over several representative methods of recommender systems and multi-task learning. Through extensive experiments on real-world datasets, we demonstrate that MKR achieves substantial gains in movie, book, music, and news recommendation, over state-of-the-art baselines. MKR is also shown to be able to maintain a decent performance even if user-item interactions are sparse.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191