A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2016; you can also visit the original URL.
The file type is application/pdf
.
Filters
Sparse Ensemble Learning for Concept Detection
2012
IEEE transactions on multimedia
This work presents a novel sparse ensemble learning scheme for concept detection in videos. ...
The resultant ensemble model is, therefore, sparse, in the way that only a small number of efficient classifiers in the ensemble will fire on a testing sample. ...
The focus of this work is to develop a sparse ensemble learning method for concept detection. ...
doi:10.1109/tmm.2011.2168198
fatcat:yg5dvk75qvgilgsxvdrslblguu
TRECVid 2013 Semantic Video Concept Detection by NTT-MD-DUT
2013
TREC Video Retrieval Evaluation
For the second aspect, we followed the subspace partition based framework we proposed in our last year work and to balance the precision and efficiency, we proposed a sparse soft-clustering method for ...
ensemble learning, which can get the optimal replication parameter. ...
Most of efforts of current systems on TRECVid Semantic Indexing (SIN) task are focusing on the above two issues [2] [3] [4] [8] , and many powerful image features and advanced classifying schemes have ...
dblp:conf/trecvid/SunSTLGL13
fatcat:bt6b2jq5v5fqtmuxzbi5bfsy6e
PECOS: Prediction for Enormous and Correlated Output Spaces
[article]
2022
arXiv
pre-print
We propose a three phase framework for PECOS: (i) in the first phase, PECOS organizes the output space using a semantic indexing scheme, (ii) in the second phase, PECOS uses the indexing to narrow down ...
In this paper, we propose the Prediction for Enormous and Correlated Output Spaces (PECOS) framework, a versatile and modular machine learning framework for solving prediction problems for very large output ...
Acknowledgement We thank Amazon for supporting this work. We also thank Lexing Ying, Philip Etter, and Tavor Baharav for providing feedback on the manuscript. ...
arXiv:2010.05878v2
fatcat:ynuop7uaurconjopoyfi232thu
One Blade for One Purpose: Advancing Math Information Retrieval using Hybrid Search
2023
Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval
Our hybrid search outperforms the previous state-of-the-art math IR system while eliminating efficiency bottlenecks. Our system is available at https://github.com/approach0/pya0. ...
To this end, we propose MABOWDOR, a Math-Aware Bestof-Worlds Domain Optimized Retriever, which has an unsupervised structure search component, a dense retriever, and optionally a sparse retriever on top ...
75] , sparse lexical weight vectors can be learned to utilize inverted indexes and existing optimizations [7, 33, 34, 36, 83] , and using better pretraining to encode a stronger single-dense representation ...
doi:10.1145/3539618.3591746
fatcat:rojzhmxsbjdnff5eqwptwmiagq
A Neural Corpus Indexer for Document Retrieval
[article]
2023
arXiv
pre-print
To this end, we propose Neural Corpus Indexer (NCI), a sequence-to-sequence network that generates relevant document identifiers directly for a designated query. ...
Current state-of-the-art document retrieval solutions mainly follow an index-retrieve paradigm, where the index is hard to be directly optimized for the final retrieval target. ...
From neural re-ranking to neural ranking: Learning a sparse representation for inverted index-
ing. ...
arXiv:2206.02743v3
fatcat:ac2azatiz5g3nemelwo5btku6u
End-to-End Retrieval with Learned Dense and Sparse Representations Using Lucene
[article]
2023
arXiv
pre-print
The bi-encoder architecture provides a framework for understanding machine-learned retrieval models based on dense and sparse vector representations. ...
., inverted indexes, HNSW indexes, and toolkits for neural inference), often knitted together in complex architectures. ...
semantics, or sparse vectors where the dimensions are defined by the vocabulary; • whether the vectors are generated by a machine-learned model (typically in a supervised manner) or via some heuristic ...
arXiv:2311.18503v1
fatcat:jhzbdlijcbdefdq3iwi2fgjlma
Domain Adaptation of Multilingual Semantic Search – Literature Review
[article]
2024
arXiv
pre-print
We also explore the possibilities of combining multilingual semantic search with domain adaptation approaches for dense retrievers in a low-resource setting. ...
We developed a new typology to cluster domain adaptation approaches based on the part of dense textual information retrieval systems, which they adapt, focusing on how to combine them efficiently. ...
However, while integrating sparse retrieval into dense retrieval systems enhances their term-matching capabilities, this approach requires maintaining two indexing systems simultaneously, which is often ...
arXiv:2402.02932v1
fatcat:h3xylj6dezdv5lgvqzrcbygggq
A Review on Automated Disease Diagnosis Techniques
2017
IJARCCE
In shallow learning methods, which make use of patient details from hospital records with structured fields, they can focus on only a single or a few diseases. ...
In this paper, a survey of different techniques for automatic disease diagnosis is done. ...
[3] proposed a disease inference system from health related questions via sparse deep learning. ...
doi:10.17148/ijarcce.2017.63187
fatcat:bxgqlbcyyzc5tjstmb6xdc6ogi
Naver Labs Europe @ TREC Deep Learning 2020
2020
Text Retrieval Conference
While the track comprises 4 tasks in total (document and passage (re-)ranking), we only focused on the passage full ranking task, for which the goal is to retrieve and rank a set of 1000 passages directly ...
This paper describes our participation to the 2020 TREC Deep Learning challenge. ...
Table 1 . 1 Performance of first-stage semantic retrieval, on MSMARCO dev set (sparse labels) and TREC Deep Learning 2019 test set. ...
dblp:conf/trec/FormalPFC20
fatcat:xlmjexqn3bdztd2vkcynbx7nyu
Combining Generative/Discriminative Learning for Automatic Image Annotation and Retrieval
2012
International Journal of Intelligence Science
Since the framework combines the advantages of generative and discriminative learning, it can predict semantic annotation precisely for unseen images. ...
Furthermore, we propose a hybrid framework which employs continuous PLSA to model visual features of images in generative learning stage and uses ensembles of classifier chains to classify the multi-label ...
Although the efficiency and accuracy of asymmetric learning approach are quite good, we believe that using multi-label learning to learn semantic concepts is more appropriate to solve the problem caused ...
doi:10.4236/ijis.2012.23008
fatcat:y7mhhjarejhlpleycxh6fsdwuu
Video anomaly detection and localization using hierarchical feature representation and Gaussian process regression
2015
2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
A novel inference method for computing the likelihood of an observed interaction is also proposed. ...
nearby sparse spatio-temporal interest points. ...
We employ a two-phase strategy for computational efficiency. It begins with partitioning an ensemble space into n r 3D subregions. ...
doi:10.1109/cvpr.2015.7298909
dblp:conf/cvpr/ChengCF15
fatcat:47pnfgr5sncarnsomsfofnebaq
Ensemble Learning with LDA Topic Models for Visual Concept Detection
[chapter]
2012
Multimedia - A Multidisciplinary Approach to Complex Issues
Recently, we proposed a localized multiple kernel learning method for realistic human action recognition based on multiple features (Song et al., 2011) , and sparse ensemble learning for visual concept ...
detection (Tang et al., 2012) by exploiting a sparse non-negative matrix factorization process to for ensemble construction and fusion. ...
doi:10.5772/37716
fatcat:34ko6xuaqbbhldqnwxvpslrak4
PECOS
2022
Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
Specifically, PECOS eases complicated semantic indexing for organizing enormous output spaces, thereby efficiently training models and deriving predictions by magnitude orders on correlated output labels ...
By way of real-world examples, the attendees will learn how to efficiently train large-scale machine learning models for enormous output spaces, and obtain predictions in less than 1 millisecond for a ...
His work mainly focuses on large-scale machine learning methods for various industrial applications such as semantic matching and recommendation systems in e-commerce. ...
doi:10.1145/3534678.3542629
fatcat:gbmzlzif2nggfors4yfirglixm
2018 Index IEEE Transactions on Knowledge and Data Engineering Vol. 30
2019
IEEE Transactions on Knowledge and Data Engineering
., þ, TKDE July 2018 1386-1402 Databases Ensemble Learning for Multi-Type Classification in Heterogeneous Networks. ...
., þ, TKDE Aug. 2018 1440-1453 249-262
SUBJECT INDEX
A
Advertising data processing
Bidding Machine: Learning to Bid for Directly Optimizing Profits in
Display Advertising. ...
doi:10.1109/tkde.2018.2882359
fatcat:asiids266jagrkx5eac6higrlq
Image Semantic Segmentation Method Based on Deep Learning in UAV Aerial Remote Sensing Image
2022
Mathematical Problems in Engineering
Therefore, an image semantic segmentation method based on deep learning in UAV aerial remote sensing images is proposed. ...
Bottleneck layer with 1 × 1 convolution is introduced to build ISegNet network model, and pooling index and convolution are used to fuse semantic information and image features. ...
Figure 5 : 5 Figure 5: Ensemble learning process based on multiple classifiers.
Figure 4 : 4 Figure 4: Maxpooling index and upsampling. ...
doi:10.1155/2022/5983045
fatcat:r4oljy3gjrgdpofokgqwcd3yqm
« Previous
Showing results 1 — 15 out of 5,524 results