Filters








10,104 Hits in 5.3 sec

Asymmetric Tri-training for Debiasing Missing-Not-At-Random Explicit Feedback [article]

Yuta Saito
2020 arXiv   pre-print
In most real-world recommender systems, the observed rating data are subject to selection bias, and the data are thus missing-not-at-random.  ...  Moreover, we show that the proposed meta-learning method is robust to these issues and can facilitate in developing effective recommendations from biased explicit feedback.  ...  It is difficult to address the effect of selection bias of real-world recommender systems when the MCAR data is unavailable.  ... 
arXiv:1910.01444v6 fatcat:earnyk432veajbpej5zj6nh45e

An Adaptive Boosting Technique to Mitigate Popularity Bias in Recommender System [article]

Ajay Gangwar, Shweta Jain
2021 arXiv   pre-print
The observed ratings in most recommender systems are subjected to popularity bias and are thus not randomly missing.  ...  In the literature, several fair algorithms have been proposed which mainly focused on improving the accuracy of the recommendation system.  ...  RELATED WORKS Fairness in recommender systems has been the center of discussion for a long time since these systems suffer from different types of biases present in the data.  ... 
arXiv:2109.05677v1 fatcat:wkgc3ouwwbdvdgxkh3cekozq2i

Causal Learning for Socially Responsible AI [article]

Lu Cheng, Ahmadreza Mosallanezhad, Paras Sheth, Huan Liu
2022 arXiv   pre-print
We begin by examining the seven CL tools to enhance the social responsibility of AI, then review how existing works have succeeded using these tools to tackle issues in developing SRAI such as fairness  ...  Acknowledgements This material is based upon work supported by, or in part by, the U.S. Army Research Laboratory and the U.S.  ...  Bias Mitigation AI systems can be biased due to hidden or neglected biases in the data, algorithms, and user interaction.  ... 
arXiv:2104.12278v2 fatcat:hk42masibjai7pnv5y2t7gj7ja

Formation of Pseudo-Terminal Restriction Fragments, a PCR-Related Bias Affecting Terminal Restriction Fragment Length Polymorphism Analysis of Microbial Community Structure

M. Egert, M. W. Friedrich
2003 Applied and Environmental Microbiology  
These peaks were called "pseudo-T-RFs" since they can be detected as terminal fluorescently labeled fragments in T-RFLP analysis but do not represent the primary terminal restriction site as indicated  ...  Digestion of amplicons with the single-strand-specific mung bean nuclease prior to T-RFLP analysis completely eliminated pseudo-T-RFs.  ...  We thank Bianca Wagner for excellent technical assistance, and we thank Gesche Braker and Tillmann Lueders for data on the in vitro T-RF formation pattern of cloned functional genes.  ... 
doi:10.1128/aem.69.5.2555-2562.2003 pmid:12732521 pmcid:PMC154551 fatcat:uo3zvxhkkffzpifq3ry4r4zmcm

SocialTransfer

Suman Deb Roy, Tao Mei, Wenjun Zeng, Shipeng Li
2012 Proceedings of the 20th ACM international conference on Multimedia - MM '12  
It is characterized by two key components: 1) a topic space learned in real time from social streams via Online Streaming Latent Dirichlet Allocation (OSLDA), and 2) a real-time cross-domain graph spectra  ...  In this paper, we investigate how to build a mutual connection among the disparate social media on the Internet, using which cross-domain media recommendation can be realized.  ...  Let us consider the simple case of two topic clusters A in 1 and A in 2 , such that A in = A in 1 ∪ A in 2 denotes the set of labeled bias instances.  ... 
doi:10.1145/2393347.2393437 dblp:conf/mm/RoyMZL12 fatcat:k2uitwucjvdhll7etnc3gy7a2i

Integration of Ultra-Low Volume Pneumatic Microfluidics with a Three-Dimensional Electrode Network for On-Chip Biochemical Sensing

Saurabh Tomar, Charlotte Lasne, Sylvain Barraud, Thomas Ernst, Carlotta Guiducci
2021 Micromachines  
It eliminates the need for post-CMOS processing and can scale up in numbers with the CMOS scaling.  ...  This paper reports a novel miniaturized pseudo reference electrode (RE) design for biasing Ion Sensitive Field Effect Transistors (ISFETs).  ...  The need for post-CMOS processing chips is eliminated by the combination of conformal pseudo-RE in microfluidics and TPV.  ... 
doi:10.3390/mi12070762 fatcat:uvrqkritifbn5ozzovd6wfdgmm

Web Description Logic Rule Generation And Other Machine Learning Algorithms – A Comparative Study

Revathi S
2020 International Journal of Advanced Trends in Computer Science and Engineering  
In order to emphasize on the security and privacy parameters, a Web Description Logic Rule Generation (WDLRG) algorithm has been recommended for identifying and evaluating vulnerable users as well as eliminating  ...  OSN users are so engrossed in building social circle that they hardly pay attention to any security related issues.  ...  In order to emphasize on the security and privacy parameters, a Web Description Logic Rule Generation (WDLRG) algorithm has been recommended for identifying and evaluating vulnerable users as well as eliminating  ... 
doi:10.30534/ijatcse/2020/140922020 fatcat:ajjat7h5obcr7ohfyrfd7dphpu

Bias and Debias in Recommender System: A Survey and Future Directions [article]

Jiawei Chen, Hande Dong, Xiang Wang, Fuli Feng, Meng Wang, Xiangnan He
2021 arXiv   pre-print
This makes various biases widely exist in the data, including but not limited to selection bias, position bias, exposure bias, and popularity bias.  ...  In this paper, we first summarize seven types of biases in recommendation, along with their definitions and characteristics.  ...  of seven types of biases in recommendation and the bias amplification in loop.  ... 
arXiv:2010.03240v2 fatcat:6fticc3otndsra2whs5e4nrdpi

Indoor Localization using Data Augmentation via Selective Generative Adversarial Networks

Wafa Njima, Marwa Chafii, Arsenia Chorti, Raed M. Shubair, H. Vincent Poor
2021 IEEE Access  
The developed model utilizes semi-supervised learning in order to predict the pseudo-labels of the generated RSSIs.  ...  In this paper, we propose generative adversarial networks for RSSI data augmentation which generate fake RSSI data based on a small set of real collected labeled data.  ...  data and fake generated pseudo-labeled data.  ... 
doi:10.1109/access.2021.3095546 fatcat:etteyus3onf7tgaix7doras2ci

Correcting the hub occurrence prediction bias in many dimensions

Nenad Tomasev, Krisztian Buza, Dunja Mladenic
2016 Computer Science and Information Systems  
This study examines the nature of the instance selection bias in intrinsically high-dimensional data.  ...  Our experiments reveal that different instance selection strategies bias the predictions of the behavior of hub-points in high-dimensional data in different ways.  ...  Hubness has first been reported in music retrieval and recommendation systems [1] , where it is still an important issue [12] [15] .  ... 
doi:10.2298/csis140929039t fatcat:c5veelea4bb2ho7lgw6btmzu4q

Coarse2Fine: Fine-grained Text Classification on Coarsely-grained Annotated Data [article]

Dheeraj Mekala, Varun Gangal, Jingbo Shang
2021 arXiv   pre-print
Instead of asking for new fine-grained human annotations, we opt to leverage label surface names as the only human guidance and weave in rich pre-trained generative language models into the iterative weak  ...  Our framework uses the fine-tuned generative models to sample pseudo-training data for training the classifier, and bootstraps on real unlabeled data for model refinement.  ...  The research was sponsored in part by National Science Foundation Convergence Accelerator under award OIA-2040727 as well as generous gifts from Google, Adobe, and Teradata.  ... 
arXiv:2109.10856v1 fatcat:xcsqk4vzgvhxpdfeprqaiozbxi

DebtFree: Minimizing Labeling Cost in Self-Admitted Technical Debt Identification using Semi-Supervised Learning [article]

Huy Tu, Tim Menzies
2022 arXiv   pre-print
In mode1, when the existing training data is unlabeled, DebtFree starts with an unsupervised learner to automatically pseudo-label the programming comments in the training data.  ...  Specifically, DebtFree can reduce the labeling effort by 99% in mode1 (unlabeled training data), and up to 63% in mode2 (labeled training data) while improving the current active learner's F1 relatively  ...  In setting 1, or DebtFree(0), the best combination is Pseudo-Labeling (via CLA) with our proposed active learner, Falcon.  ... 
arXiv:2201.10592v1 fatcat:kiqgn3q54vfk3f3uzaudegmwpu

You Think You've Got Trivials?

Shlomo S. Sawilowsky
2003 Journal of Modern Applied Statistical Methods  
In this article, clarifications are given regarding what should be simulated to determine the possible effects of piecemeal publishing trivial effect sizes.  ...  Note, however, that no system or phenomenon was simulated.  ...  a uniform pseudo-random number generator and computing the mean, but usually there is little point in doing so.)  ... 
doi:10.22237/jmasm/1051748460 fatcat:svbrrg2udbchnaejrn2d2l3zki

A Linear Programming Approach to Multiple Instance Learning

2020 Turkish Journal of Electrical Engineering and Computer Sciences  
In MIL, objects are represented by a bag of instances instead of a single instance 4 and class labels are provided only for the bags.  ...  Each instance of a bag is mapped to a pseudo-class membership estimate 8 and these estimates are aggregated to obtain the bag-level class membership in an optimization framework.  ...  We assess the pseudo-membership values of instances to find bag-level estimates, 5 not for instance labeling since the actual instance labels are not known in MIL tasks.  ... 
doi:10.3906/elk-2009-144 fatcat:c77nrs7o2zhx7o6ahhbvhlv56e

Non-linear fitting with joint spatial regularization in Arterial Spin Labeling [article]

Oliver Maier, Stefan M Spann, Daniela Pinter, Thomas Gattringer, Nicole Hinteregger, Gerhard G. Thallinger, Christian Enzinger, Josef Pfeuffer, Kristian Bredies, Rudolf Stollberger
2021 arXiv   pre-print
Multi-Delay single-shot arterial spin labeling (ASL) imaging provides accurate cerebral blood flow (CBF) and, in addition, arterial transit time (ATT) maps but the inherent low SNR can be challenging.  ...  Especially standard fitting using non-linear least squares often fails in regions with poor SNR, resulting in noisy estimates of the quantitative maps.  ...  Algorithm 1: Primal-dual algorithm for solving the TGV 2 regularized ASL parameter quantification task in every Gauss-Newton step.  ... 
arXiv:2009.05409v2 fatcat:4cfp5c26kvg35hdv4temndqcyu
« Previous Showing results 1 — 15 out of 10,104 results