Recommendation systems play a vital role to keep users engaged with personalized content in modern online platforms. Deep learning has revolutionized many research fields and there is a recent surge of interest in applying it to collaborative filtering (CF). However, existing methods compose deep learning architectures with the latent factor model ignoring a major class of CF models, neighborhood or memory-based approaches. We propose Collaborative Memory Networks (CMN), a deep architecture to unify the two classes of CF models capitalizing on the strengths of the global structure of latent factor model and local neighborhood-based structure in a nonlinear fashion. Motivated by the success of Memory Networks, we fuse a memory component and neural attention mechanism as the neighborhood component. The associative addressing scheme with the user and item memories in the memory module encodes complex user-item relations coupled with the neural attention mechanism to learn a user-item specific neighborhood. Finally, the output module jointly exploits the neighborhood with the user and item memories to produce the ranking score. Stacking multiple memory modules together yield deeper architectures capturing increasingly complex user-item relations. Furthermore, we show strong connections between CMN components, memory networks and the three classes of CF models. Comprehensive experimental results demonstrate the effectiveness of CMN on three public datasets outperforming competitive baselines. Qualitative visualization of the attention weights provide insight into the model's recommendation process and suggest the presence of higher order interactions.
Abstract. Bayesian belief nets (BNs) are often used for classification tasks -typically to return the most likely class label for each specified instance. Many BN-learners, however, attempt to find the BN that maximizes a different objective function -viz., likelihood, rather than classification accuracy -typically by first learning an appropriate graphical structure, then finding the parameters for that structure that maximize the likelihood of the data. As these parameters may not maximize the classification accuracy, "discriminative parameter learners" follow the alternative approach of seeking the parameters that maximize conditional likelihood (CL), over the distribution of instances the BN will have to classify. This paper first formally specifies this task, shows how it extends standard logistic regression, and analyzes its inherent sample and computational complexity. We then present a general algorithm for this task, ELR, that applies to arbitrary BN structures and that works effectively even when given incomplete training data. Unfortunately, ELR is not guaranteed to find the parameters that optimize conditional likelihood; moreover, even the optimal-CL parameters need not have minimal classification error. This paper therefore presents empirical evidence that ELR produces effective classifiers, often superior to the ones produced by the standard "generative" algorithms, especially in common situations where the given BN-structure is incorrect.
The seamless integration of low-power, miniaturised, invasive/non-invasive lightweight sensor nodes have contributed to the development of a proactive and unobtrusive Wireless Body Area Network (WBAN). A WBAN provides long-term health monitoring of a patient without any constraint on his/her normal dailylife activities. This monitoring requires the low-power operation of invasive/non-invasive sensor nodes. In other words, a power-efficient Medium Access Control (MAC) protocol is required to satisfy the stringent WBAN requirements, including low-power consumption. In this paper, we first outline the WBAN requirements that are important for the design of a low-power MAC protocol. Then we study low-power MAC protocols proposed/investigated for a WBAN with emphasis on their strengths and weaknesses. We also review different power-efficient mechanisms for a WBAN. In addition, useful suggestions are given to help the MAC designers to develop a low-power MAC protocol that will satisfy the stringent requirements.
The CPGTs significantly improved standardization, efficiency, and efficacy of cancer pain therapy in China. In a country where clinical pharmacy is still developing, this is a valuable service model that may enhance cancer treatment capacity and efficacy while promoting recognition of the clinical pharmacy profession.
Abstract-In visual tracking, holistic and part-based representations are both popular choices to model target appearance. The former is known for great efficiency and convenience while the latter for robustness against local appearance or shape variations. Based on non-negative matrix factorization (NMF), we propose a novel visual tracker that takes advantage of both groups. The idea is to model the target appearance by a non-negative combination of non-negative components learned from examples observed in previous frames. To adjust NMF to the tracking context, we include sparsity and smoothness constraints in addition to the non-negativity one. Furthermore, an online iterative learning algorithm, together with a proof of convergence, is proposed for efficient model updating. Putting these ingredients together with a particle filter framework, the proposed tracker, Constrained Online Non-negative Matrix Factorization (CONMF), achieves robustness to challenging appearance variations and non-trivial deformations while runs in real time. We evaluate the proposed tracker on various benchmark sequences containing targets undergoing large variations in scale, pose or illumination. The robustness and efficiency of CONMF is validated in comparison with several state-of-the-art trackers.
The integration of wireless sensor network (WSN) and cognitive radio (CR) technology enables a new paradigm of communication: cognitive radio sensor networks (CRSN). The existing WSN clustering algorithm cannot consider the advantage of channel resource brought by CR function in CRSN, and the CR network (CRN) clustering algorithm is designed based on the infinite energy nodes; thus both algorithms cannot operate with energy efficiency in CRSN. The paper proposes a low-energy adaptive uneven clustering hierarchy for CRSN, which can not only consider the advantage of the channel resource in reducing the energy consumption but also employ uneven clustering method for balancing the energy consumption among the cluster heads under multiple hops transmission means. Simulation results show that compared with the existing several typical clustering algorithms including WSN and CRSN clustering algorithms, low-energy adaptive clustering hierarchy (LEACH), HEED, energy-efficient unequal clustering (EEUC), cognitive LEACH (CogLEACH), and distributed spectrum-aware clustering (DSAC), the proposed algorithm can not only efficiently balance the energy consumption among cluster heads and network load in CRSN but also remarkably prolong the network lifetime.
This paper investigates linear soft combination schemes for cooperative spectrum sensing in cognitive radio networks. We propose two weight‐setting strategies under different basic optimality criteria to improve the overall sensing performance in the network. The corresponding optimal weights are derived, which are determined by the noise power levels and the received primary user signal energies of multiple cooperative secondary users in the network. However, to obtain the instantaneous measurement of these noise power levels and primary user signal energies with high accuracy is extremely challenging. It can even be infeasible in practical implementations under a low signal‐to‐noise ratio regime. We therefore propose reference data matrices to scavenge the indispensable information of primary user signal energies and noise power levels for setting the proposed combining weights adaptively by keeping records of the most recent spectrum observations. Analyses and simulation results demonstrate that the proposed linear soft combination schemes outperform the conventional maximal ratio combination and equal gain combination schemes and yield significant performance improvements in spectrum sensing.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
334 Leonard St
Brooklyn, NY 11211
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.