A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Filters
Proceedings of the IJCAI 2017 Workshop on Learning in the Presence of Class Imbalance and Concept Drift (LPCICD'17)
[article]
2017
arXiv
pre-print
Preserved Fulltext
For example, drift detection algorithms based on the traditional classification error may be sensitive to the imbalanced degree and become less effective; and class imbalance techniques need to be adaptive ...
Concept drift is a change in the underlying distribution of the problem, and is a significant issue specially when learning from data streams. It requires learners to be adaptive to dynamic changes. ...
In this paper, we focus on an oversampling method based on shapelets extraction for imbalanced time series classification. ...
arXiv:1707.09425v1
fatcat:zm6endqlbzfcldpxlreu72t4ya
On Handling Catastrophic Forgetting for Incremental Learning of Human Physical Activity on the Edge
[article]
2023
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2023; you can also visit the original URL.
The file type is application/pdf.
In particular, we consider the practical challenge of extremely limited data during the incremental learning process on edge, where catastrophic forgetting is required to be handled in a practical way. ...
In particular, with recent machine learning (ML) techniques, the recognition task has been largely investigated by companies and integrated into their products for customers. ...
Similarly, GDumb [24] optimized the class prototype selection via Greedy Search; Instead of building up a memory replay for old-class data, Generative Replay [29] learns a generative model and a solver ...
arXiv:2302.09310v1
fatcat:wsz7ytosjndwbbjxsmnyjkhtna
LEVERAGING TEMPORAL SUBSEQUENCES FOR TIME-SERIES CLASSIFICATION
[article]
2020
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
First, the problem of highly imbalanced time-series classification using shapelets is investigated. ...
Thus, building more robust time-series classification models is imperative. ...
The PSOD is a search-based greedy procedure to extract unique shapelets and identify orders among the selected shapelets. ...
doi:10.34944/dspace/3481
fatcat:mobi2mkrcngorevvxdrfnzj6hy
Application of Machine Learning in Wireless Networks: Key Techniques and Open Issues
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
, open data sets and platforms for researchers, theoretical guidance for ML implementation and so on are discussed. ...
As a key technique for enabling artificial intelligence, machine learning (ML) is capable of solving complex problems without explicit programming. ...
The FIS input consists of the magnitude of hysteresis margin and the errors of several KPIs like CDR, and ε-greedy EEP is adopted for each rule in the rule set. ...
arXiv:1809.08707v2
fatcat:6tnzliwthfehrpuxpmm45hs4vq
Recent Advances in Object Detection in the Age of Deep Convolutional Neural Networks
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2018 pre-print arXiv:1809.03193v1 fatcat:regs42ec3jbsjop634h3nvkxba
This strong interest can be explained not only by the importance this task has for many applications but also by the phenomenal advances in this area since the arrival of deep convolutional neural networks ...
Object detection-the computer vision task dealing with detecting instances of objects of a certain class (e.g., 'car', 'plane', etc.) in images-attracted a lot of attention from the community during the ...
The goal of this dataset is to encourage the development of algorithms for 'in the wild' data featuring large numbers of imbalanced, one-grained, categories. ...
arXiv:1809.03193v2
fatcat:wj2bu3ewvbdq5fjyvrbqewpxzu
Personalized mobile physical activity recognition
2013
Proceedings of the 17th annual international symposium on International symposium on wearable computers - ISWC '13
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Therefore, the focus of this thesis is on the development of new approaches for physical activity recognition and intensity estimation, which extend the applicability of such systems. ...
This thesis follows the goal to develop a mobile, personalized physical activity monitoring system applicable for everyday life scenarios. ...
Moreover, Rakthanmanon and Keogh [129] use the PAMAP dataset to evaluate their proposed time series shapelet discovery algorithm, and to demonstrate that shapelets can also be used as a high accuracy ...
doi:10.1145/2493988.2494349
dblp:conf/iswc/ReissS13
fatcat:mrbwnktnkncd7colng2lievxhu
A scalable machine learning system for anomaly detection in manufacturing
2024
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2024; you can also visit the original URL.
The file type is application/pdf.
He served as an invaluable sparring partner and lent his unconditional support throughout the entirety of my doctoral studies. ...
transformation algorithm (ma-genta) vs. the margin-based greedy shapelet search (blue) [7] . . . . . . . . . . . 67 Changes to the query set of the adaptive search system over multiple relevance feedback ...
classification of time series objects. ...
doi:10.17877/de290r-24118
fatcat:d5exu55zeffupju2p6pzg4ikve
Machine learning approaches for time series problems
2022
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Every sequence of data where the order in terms of time matters consists of a time series. This type of data has unique properties. ...
Time series data emerge from the dynamics of complex systems and are high dimensional. ...
Acknowledgements First of all, I would like to express my sincere gratitude to my supervisor, Prof. Dimitris Vrakas for all the support and guidance he has offered me throughout the thesis. ...
doi:10.26262/heal.auth.ir.340363
fatcat:fqggq54crnaxpmaixk4go7kraa
Distributed analysis of vertically partitioned sensor measurements under communication constraints
2017
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
They measure, for instance, parameters of production processes, environmental conditions of transported goods, energy consumption of smart homes, traffic volume, air pollution and water consumption, or ...
In 2010, already 12.5 billion devices were connected to the IoT, a number about twice as large as the world's population at that time. ...
Returned is the best labeling found over all starts of the different greedy searches. ...
doi:10.17877/de290r-17839
fatcat:qlo3l2mgrrh4dp3gwaktmcyfm4
From pixels to people : recovering location, shape and pose of humans in images
[article]
2022
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
We contribute a method for recovering 3D human shape and pose, which marries the advantages of learning-based and modelbased approaches. ...
Endowing machines with the ability to perceive people from visual data is an immense scientific challenge with a high degree of direct practical relevance. ...
Acknowledgements This thesis took much more time to produce than it should have, and an exhaustive list of people whom I am grateful to is accordingly very long and would fill several pages. ...
doi:10.22028/d291-36605
fatcat:efcywrb5jzhl7kymuorpf3bm3i
Analysis and improvement of the visual object detection pipeline
[article]
2017
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Our deep network outperforms all previous neural networks for pedestrian detection by a large margin, even without using additional training data. ...
As a side-effect we publish new, better localised annotations for the Caltech pedestrian benchmark. We analyse detection proposals as a preprocessing step for object detectors. ...
We thus employ a greedy search scheme, which runs the minimization for k = 1, fixes (p 1 , θ 1 ), and tries all possible second locations for k = 2. This strategy is repeated for each k > 2. ...
doi:10.22028/d291-26774
fatcat:q5dxn7afhncqhlypxt3nthvkby