Filters








3,768 Hits in 8.3 sec

With a Little Help from My Friends: Nearest-Neighbor Contrastive Learning of Visual Representations [article]

Debidatta Dwibedi, Yusuf Aytar, Jonathan Tompson, Pierre Sermanet, Andrew Zisserman
2021 arXiv   pre-print
Our method, Nearest-Neighbor Contrastive Learning of visual Representations (NNCLR), samples the nearest neighbors from the dataset in the latent space, and treats them as positives.  ...  We find that using the nearest-neighbor as positive in contrastive losses improves performance significantly on ImageNet classification, from 71.7% to 75.6%, outperforming previous state-of-the-art methods  ...  Next we introduce our approach, Nearest-Neighbor Contrastive Learning of visual Representations (NNCLR), which proposes using nearest-neighbours (NN) as positives to improve contrastive instance discrimination  ... 
arXiv:2104.14548v2 fatcat:vlasnfthtrd2fbnzlanzqki4ae

"With a Little Help from My Friends": The Influence of Co-offenders on the Journey to Crime [article]

(:Unkn) Unknown, University, My, Jerry H. Ratcliffe
2021
This dissertation uses a dataset of official arrest records from the City of Philadelphia, PA for 2010 to 2017 (inclusive), containing 50,928 arrest records and 14,735 individual offenders with at least  ...  one arrest on their own and one arrest with a co-offender.  ...  A visual representation of this issue is presented in the two crime trips in example A of Figure 18 .  ... 
doi:10.34944/dspace/6467 fatcat:vmyumvmqtfbwtetfsl3sl6dkfa

Homelessness research: A guide for economists (and friends)

Brendan O'Flaherty
2019 Journal of Housing Economics  
Aggregate-level studies can be used more for finding out what the effects of different policies are, and individual-level studies can be used more for assessing the costs and benefits of those effects.  ...  The emphases are on the last decade, rather than earlier; and the United States, rather than the rest of the world. The approach is more idiosyncratic than encyclopedic.  ...  Often prevention programs give families information about how to access help from other agencies; those families can pass that information on to their neighbors, friends, and relatives.  ... 
doi:10.1016/j.jhe.2019.01.003 fatcat:o7ffpu77xndhpe5n6u4pjb7cye

Recommending friends and locations based on individual location history

Yu Zheng, Lizhu Zhang, Zhengxin Ma, Xing Xie, Wei-Ying Ma
2011 ACM Transactions on the Web  
Second, we measure the similarity between users in terms of their location histories and recommend to each user a group of potential friends in a GIS community.  ...  The increasing availability of location-acquisition technologies (GPS, GSM networks, etc.) enables people to log the location histories with spatio-temporal data.  ...  I tried another branch restaurant of JaioYe in Zhongguanchun when celebrating the birthday of my friend last year.  ... 
doi:10.1145/1921591.1921596 fatcat:k7ajvdt4hnem7f2vj2thxlj4su

Revisiting Contrastive Methods for Unsupervised Learning of Visual Representations [article]

Wouter Van Gansbeke, Simon Vandenhende, Stamatios Georgoulis, Luc Van Gool
2021 arXiv   pre-print
We show that learning additional invariances -- through the use of multi-scale cropping, stronger augmentations and nearest neighbors -- improves the representations.  ...  Finally, we observe that MoCo learns spatially structured representations when trained with a multi-crop strategy.  ...  In: International Conference on Learning Representations (ICLR) (2021) 17 [17] Dwibedi, D., Aytar, Y., Tompson, J., Sermanet, P., Zisserman, A.: With a little help from my friends: Nearest-neighbor  ... 
arXiv:2106.05967v3 fatcat:a5j2q5vlyzhi5ogsyekkc2l4xa

Constrained Mean Shift Using Distant Yet Related Neighbors for Representation Learning [article]

Ajinkya Tejankar, Soroush Abbasi Koohpayegani, KL Navaneet, Kossar Pourahmadi, Akshayvarun Subramanya, Hamed Pirsiavash
2021 arXiv   pre-print
We believe the learning can benefit from choosing far away neighbors that are still semantically related to the query.  ...  neighbors (NNs) of its other augmentation.  ...  With a little help from [13] Jane Bromley, Isabelle Guyon, Yann LeCun, Eduard my friends: Nearest-neighbor contrastive learning of visual Säckinger, and Roopak Shah.  ... 
arXiv:2112.04607v1 fatcat:n7g5f2obnzf6xjn6jpyjpp4ezu

Designing for effective end-user interaction with machine learning

Saleema Amershi
2011 Proceedings of the 24th annual ACM symposium adjunct on User interface software and technology - UIST '11 Adjunct  
However, we still lack a generalized understanding of how to design effective end-user interaction with machine learning.  ...  Designing for Effective End-User Interaction with Machine Learning Saleema Amershi Chair of the Supervisory Committee: Associate Professor James A.  ...  ACKNOWLEDGEMENTS I would like to begin by thanking my wonderful advisor James Fogarty for his guidance and endless support throughout my graduate career.  ... 
doi:10.1145/2046396.2046416 dblp:conf/uist/Amershi11 fatcat:q3ticzzymngu5i7uhdcyshmt34

Can x2vec Save Lives? Integrating Graph and Language Embeddings for Automatic Mental Health Classification [article]

Alexander Ruch
2020 arXiv   pre-print
Graph and language distances to a suicide support group have little correlation (h̊o̊ < 0.23), implying the two models are not embedding redundant information.  ...  predictions of potentially suicidal individuals shows the integrated model could classify such individuals even if they are positioned far from the support group.  ...  The author is also grateful for assistance and feedback from Seunghyun Kim, Lillyan Pan, Hannah Lee, James Zou, Jeffrey Tsang, Bryan Min, Juliana Hong, and Cornell University's Social Dynamics Laboratory  ... 
arXiv:2001.01126v1 fatcat:2ts4u7lyuvblfiryp6yzhihkqm

Does BERT Make Any Sense? Interpretable Word Sense Disambiguation with Contextualized Embeddings [article]

Gregor Wiedemann, Steffen Remus, Avi Chawla, Chris Biemann
2019 arXiv   pre-print
We introduce a simple but effective approach to WSD using a nearest neighbor classification on CWEs.  ...  ., 2019) are a major recent innovation in NLP. CWEs provide semantic vector representations of words depending on their respective context.  ...  Acknowledgments This work was funded by BWFG Hamburg in the Forum 4.0 project, by DFG in the JOIN-T 2 project and by DAAD in form of a WISE stipend.  ... 
arXiv:1909.10430v2 fatcat:jo5thzssjzgvtlo3pwafoxqyjq

Multi-class object localization by combining local contextual interactions

Carolina Galleguillos, Brian McFee, Serge Belongie, Gert Lanckriet
2010 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition  
Context, obtained from the object's nearby image data, image annotations and the presence and location of other objects, can help to disambiguate appearance inputs in recognition tasks.  ...  The use of this technology can be advantageous in guiding a blind user to recognize objects in real time and augmenting the ability of search engines to permit searches based on image content.  ...  ., support vector machines), with relatively little attention given to the optimization of nearest neighbor classifiers.  ... 
doi:10.1109/cvpr.2010.5540223 dblp:conf/cvpr/GalleguillosMBL10 fatcat:asm6h3hix5fufkipt56voasqcu

VideoSET: Video Summary Evaluation through Text [article]

Serena Yeung, Alireza Fathi, Li Fei-Fei
2014 arXiv   pre-print
Given a video summary, a text representation of the video summary is first generated, and an NLP-based metric is then used to measure its semantic distance to ground-truth text summaries written by humans  ...  We also release text annotations and ground-truth text summaries for a number of publicly available video datasets, for use by the computer vision community.  ...  Acknowledgements This research is partially supported by an ONR MURI grant and an Intel gift, and a Stanford Graduate Fellowship to S.Y.  ... 
arXiv:1406.5824v1 fatcat:yq6qeq7abzhgpcwqxxtufl74qi

Outcome-Explorer: A Causality Guided Interactive Visual Interface for Interpretable Algorithmic Decision Making [article]

Md Naimul Hoque, Klaus Mueller
2021 arXiv   pre-print
To demonstrate our method we developed Outcome Explorer, a causality guided interactive interface, and evaluated it by conducting think-aloud sessions with three expert users and a user study with 18 non-expert  ...  The widespread adoption of algorithmic decision-making systems has brought about the necessity to interpret the reasoning behind these decisions.  ...  How am I different than my friend?  ... 
arXiv:2101.00633v2 fatcat:kqrh7mrl6rcnvbcdqqnl6nr2ku

Sentiment Analysis on Multi-View Social Data [chapter]

Teng Niu, Shiai Zhu, Lei Pang, Abdulmotaleb El Saddik
2016 Lecture Notes in Computer Science  
I would also like to thank all my colleagues in the MCRLab for their suggestions and contributions throughout the research, and all my friends for their help on my campus life.  ...  In this thesis, we further conduct a comprehensive study on computational analysis of sentiment from the multi-view data.  ...  A standard way to address this problem leverages the supervised learning techniques with visual features extracted from a set of training data.  ... 
doi:10.1007/978-3-319-27674-8_2 fatcat:qwi4nsyelbhxdh64ekeq7dyscy

Human-Centered Tools for Coping with Imperfect Algorithms during Medical Decision-Making [article]

Carrie J. Cai, Emily Reif, Narayan Hegde, Jason Hipp, Been Kim, Daniel Smilkov, Martin Wattenberg, Fernanda Viegas, Greg S. Corrado, Martin C. Stumpe, Michael Terry
2019 arXiv   pre-print
One application of ML is to retrieve visually similar medical images from past patients (e.g. tissue from biopsies) to reference when making a medical decision with a new patient.  ...  In this paper, we identified the needs of pathologists when searching for similar images retrieved using a deep learning algorithm, and developed tools that empower users to cope with the search algorithm  ...  ACKNOWLEDGEMENTS We are gra teful to th e following individuals for their valuable help and feedback on this work: Trissia Brown, Michael Emmert-Buck, Isabelle Flament, Fraser Tan, Lily Peng, Craig Mermel  ... 
arXiv:1902.02960v1 fatcat:o6vfb4ifrbgwzhwqlmgp3wikou

Math That Matters: Enhancing Academic Mathematics' Impact on Society [article]

Christopher Thron, Monira Taj Elsir Hamid Ali
2021 arXiv   pre-print
In this paper we present three examples of mathematics with significant social benefit: Dmitri Bertsimas' study of diabetes using k-nearest-neighbor methodology; Development of mathematical software (MATLAB  ...  and Sage); and ongoing development of data representation and visualization software to facilitate analysis of survey data.  ...  (Readers familiar with machine learning may recognize this as a variation of the standard classification technique known as "k-nearest neighbor".)  ... 
arXiv:2103.17067v2 fatcat:uxdoagjtfjg7tjspx3yiuv7364
« Previous Showing results 1 — 15 out of 3,768 results