continualai / continual-learning-papers Goto Github PK

View Code? Open in Web Editor NEW

541.0 541.0 51.0 7.58 MB

Continual Learning papers list, curated by ContinualAI

License: MIT License

TeX 22.41% Python 0.66% HTML 76.93%

awesome-list continual-learning lifelong-learning online-learning papers

continual-learning-papers's Introduction

ContinualAI - Jekyll Website now DEPRECATED

We moved to GitBook. If you want to help us maintain the new website, please send and email to [email protected].

Continual AI is the first hub on Continual / Lifelong Deep Learning in AI! :-) The aim of the project is to provide a starting point for researchers, developers and AI enthusiasts who share an interest or are willing to learn more and/or contribute to Continual / Lifelong Learning. We are building an open-source, collaborative wiki at continualai.org as well as creating a community of CL enthusiasts! Join us today on slack! :D

How to contribute

Star the project :-)
Join our community on Slack: https://continualai.herokuapp.com/
Start making changes to the *.md files from the browser (use the 'Preview' button)
Commit the changes!

How to contribute (like a pro)

Star the project :-)
Join our community on Slack: https://continualai.herokuapp.com/
Fork the repo on GitHub and clone it locally
Enter the folder:

cd website-wiki
If you don't have gem and bundler installed:

apt-get install rubygems
gem install bundler
Install Ruby gems:

bundle install
Start Jekyll server:

jekyll serve --incremental
Now you can start making changes on the see the result in your browser at http://localhost:4000/
Make a Pull Request (with only the .md or original .html files)! :D

continual-learning-papers's People

Contributors

Stargazers

Watchers

continual-learning-papers's Issues

Adding the paper: Towards Continual Reinforcement Learning: A Review and Perspectives

Hello,

Thank you so much for these useful resources!

Maybe it's nice to add this paper to the continual learning for RL section:

Towards Continual Reinforcement Learning: A Review and Perspectives

Link: https://arxiv.org/abs/2012.13490

Javascript filters not working properly

The interactive website does not apply properly the javascript filters. The problem is connected to the interaction with the DOM. To solve this a careful check of the DOM structure is needed in order to query the correct DOM elements.

Add link to bib directly next to paper in readme

NeurIPS 2020 Zixuan [Catastrophic Forgetting]

@article{https://doi.org/10.48550/arxiv.2112.10017, doi = {10.48550/ARXIV.2112.10017}, url = {https://arxiv.org/abs/2112.10017}, author = {Ke, Zixuan and Liu, Bing and Huang, Xingchang}, keywords = {Machine Learning (cs.LG), Artificial Intelligence (cs.AI), Computer Vision and Pattern Recognition (cs.CV), Neural and Evolutionary Computing (cs.NE), FOS: Computer and information sciences, FOS: Computer and information sciences}, title = {Continual Learning of a Mixed Sequence of Similar and Dissimilar Tasks}, publisher = {arXiv}, year = {2021}, copyright = {Creative Commons Zero v1.0 Universal} }

Add link to CLsurvey codebase

We should add a link to this codebase: https://github.com/Mattdl/CLsurvey

[Thesis] Toward Generate-and-Test Algorithms for Continual Feature Discovery

http://incompleteideas.net/papers/Rahman-MSc.pdf

New CL Thesis

New CL Thesis here:

Open Set Classification for Deep Learning in Large-Scale and Continual Learning Models Continual Learning Models: https://scholarworks.rit.edu/cgi/viewcontent.cgi?article=11740&context=theses

Add references to popular benchmarks

In our Zotero group, section Benchmarks, we should add references to benchmarks like Split MNIST / Omniglot and so on.

Add papers and PhD thesis

Thank you for the great papers library. Can you please add the following? Thank you!

Thesis:

@phdthesis{DBLP:phd/hal/Belouadah21,
  author    = {Eden Belouadah},
  title     = {Large-scale deep class-incremental learning. (Apprentissage incr{\'{e}}mental
               profond {\`{a}} large {\'{e}}chelle)},
  school    = {Ecole nationale sup{\'{e}}rieure Mines-T{\'{e}}l{\'{e}}com
               Atlantique, France},
  year      = {2021},
  url       = {https://tel.archives-ouvertes.fr/tel-03478553},
  timestamp = {Wed, 26 Jan 2022 22:09:47 +0100},
  biburl    = {https://dblp.org/rec/phd/hal/Belouadah21.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

papers:

*** Memory-free:** 
@inproceedings{DBLP:conf/wacv/SlimB0O22,
  author    = {Habib Slim and
               Eden Belouadah and
               Adrian Popescu and
               Darian M. Onchis},
  title     = {Dataset Knowledge Transfer for Class-Incremental Learning without
               Memory},
  booktitle = {{IEEE/CVF} Winter Conference on Applications of Computer Vision, {WACV}
               2022, Waikoloa, HI, USA, January 3-8, 2022},
  pages     = {3311--3320},
  publisher = {{IEEE}},
  year      = {2022},
  url       = {https://doi.org/10.1109/WACV51458.2022.00337},
  doi       = {10.1109/WACV51458.2022.00337},
  timestamp = {Thu, 17 Feb 2022 14:51:17 +0100},
  biburl    = {https://dblp.org/rec/conf/wacv/SlimB0O22.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

*** Review Papers and Books:**
@article{DBLP:journals/corr/abs-2202-00386,
  author    = {Umang Aggarwal and
               Adrian Popescu and
               Eden Belouadah and
               C{\'{e}}line Hudelot},
  title     = {A Comparative Study of Calibration Methods for Imbalanced Class Incremental
               Learning},
  journal   = {CoRR},
  volume    = {abs/2202.00386},
  year      = {2022},
  url       = {https://arxiv.org/abs/2202.00386},
  eprinttype = {arXiv},
  eprint    = {2202.00386},
  timestamp = {Wed, 09 Feb 2022 15:43:35 +0100},
  biburl    = {https://dblp.org/rec/journals/corr/abs-2202-00386.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

*** Review Papers and Books:**
@article{DBLP:journals/nn/BelouadahPK21,
  author    = {Eden Belouadah and
               Adrian Popescu and
               Ioannis Kanellos},
  title     = {A comprehensive study of class incremental learning algorithms for
               visual tasks},
  journal   = {Neural Networks},
  volume    = {135},
  pages     = {38--54},
  year      = {2021},
  url       = {https://doi.org/10.1016/j.neunet.2020.12.003},
  doi       = {10.1016/j.neunet.2020.12.003},
  timestamp = {Tue, 01 Jun 2021 09:59:41 +0200},
  biburl    = {https://dblp.org/rec/journals/nn/BelouadahPK21.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}


*** Memory-free:**
@inproceedings{DBLP:conf/eccv/BelouadahP18,
  author    = {Eden Belouadah and
               Adrian Popescu},
  editor    = {Laura Leal{-}Taix{\'{e}} and
               Stefan Roth},
  title     = {DeeSIL: Deep-Shallow Incremental Learning},
  booktitle = {Computer Vision - {ECCV} 2018 Workshops - Munich, Germany, September
               8-14, 2018, Proceedings, Part {II}},
  series    = {Lecture Notes in Computer Science},
  volume    = {11130},
  pages     = {151--157},
  publisher = {Springer},
  year      = {2018},
  url       = {https://doi.org/10.1007/978-3-030-11012-3\_11},
  doi       = {10.1007/978-3-030-11012-3\_11},
  timestamp = {Fri, 28 Aug 2020 13:32:32 +0200},
  biburl    = {https://dblp.org/rec/conf/eccv/BelouadahP18.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

*** Rehearsal methods:**
@inproceedings{DBLP:conf/iccv/BelouadahP19,
  author    = {Eden Belouadah and
               Adrian Popescu},
  title     = {{IL2M:} Class Incremental Learning With Dual Memory},
  booktitle = {2019 {IEEE/CVF} International Conference on Computer Vision, {ICCV}
               2019, Seoul, Korea (South), October 27 - November 2, 2019},
  pages     = {583--592},
  publisher = {{IEEE}},
  year      = {2019},
  url       = {https://doi.org/10.1109/ICCV.2019.00067},
  doi       = {10.1109/ICCV.2019.00067},
  timestamp = {Fri, 28 Aug 2020 13:32:30 +0200},
  biburl    = {https://dblp.org/rec/conf/iccv/BelouadahP19.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}


*** Rehearsal methods:**
@inproceedings{DBLP:conf/wacv/BelouadahP20,
  author    = {Eden Belouadah and
               Adrian Popescu},
  title     = {ScaIL: Classifier Weights Scaling for Class Incremental Learning},
  booktitle = {{IEEE} Winter Conference on Applications of Computer Vision, {WACV}
               2020, Snowmass Village, CO, USA, March 1-5, 2020},
  pages     = {1255--1264},
  publisher = {{IEEE}},
  year      = {2020},
  url       = {https://doi.org/10.1109/WACV45572.2020.9093562},
  doi       = {10.1109/WACV45572.2020.9093562},
  timestamp = {Fri, 28 Aug 2020 13:32:27 +0200},
  biburl    = {https://dblp.org/rec/conf/wacv/BelouadahP20.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}


*** Active Learning:**
@inproceedings{DBLP:conf/eccv/BelouadahPAS20,
  author    = {Eden Belouadah and
               Adrian Popescu and
               Umang Aggarwal and
               L{\'{e}}o Saci},
  editor    = {Adrien Bartoli and
               Andrea Fusiello},
  title     = {Active Class Incremental Learning for Imbalanced Datasets},
  booktitle = {Computer Vision - {ECCV} 2020 Workshops - Glasgow, UK, August 23-28,
               2020, Proceedings, Part {VI}},
  series    = {Lecture Notes in Computer Science},
  volume    = {12540},
  pages     = {146--162},
  publisher = {Springer},
  year      = {2020},
  url       = {https://doi.org/10.1007/978-3-030-65414-6\_12},
  doi       = {10.1007/978-3-030-65414-6\_12},
  timestamp = {Wed, 07 Apr 2021 16:01:44 +0200},
  biburl    = {https://dblp.org/rec/conf/eccv/BelouadahPAS20.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}


*** Memory-free:**
@inproceedings{DBLP:conf/bmvc/BelouadahPK20,
  author    = {Eden Belouadah and
               Adrian Popescu and
               Ioannis Kanellos},
  title     = {Initial Classifier Weights Replay for Memoryless Class Incremental
               Learning},
  booktitle = {31st British Machine Vision Conference 2020, {BMVC} 2020, Virtual
               Event, UK, September 7-10, 2020},
  publisher = {{BMVA} Press},
  year      = {2020},
  url       = {https://www.bmvc2020-conference.com/assets/papers/0743.pdf},
  timestamp = {Wed, 03 Feb 2021 08:36:16 +0100},
  biburl    = {https://dblp.org/rec/conf/bmvc/BelouadahPK20.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

Add CVPR 2021 papers

[Review] Recent Advances of Continual Learning in Computer Vision: An Overview

https://arxiv.org/abs/2109.11369

Integrate other paper lists

Add link to external paper lists:

Add content from old manual lists

We need to check the bibitex contains entries we listed manually in the past and thought to be important: https://github.com/ContinualAI/wiki/blob/0bc778edeb428f98fec5f87c3fee9044d63d2858/research.rst

Anatomy of Catastrophic Forgetting: Hidden Representations and Task Semantics - ICLR 2021

Hi,

Thanks for maintaining this great repo! I believe this paper is missing from the Catastrophic Forgetting Studies:

@inproceedings{
ramasesh2021anatomy,
title={Anatomy of Catastrophic Forgetting: Hidden Representations and Task Semantics},
author={Vinay Venkatesh Ramasesh and Ethan Dyer and Maithra Raghu},
booktitle={International Conference on Learning Representations},
year={2021},
url={https://openreview.net/forum?id=LhY8QdUGSuw}
}

Move HTML+JS papers list to this project pages

We need to move HTML+JS papers list to this project pages as the Jekyll CLAI website has been just DEPRECATED in favor of GitBook.
Adjust also the README accordingly.

Add CVPR 2019 De Lange [Catastrophic Forgetting]

@article{Delange_2021, doi = {10.1109/tpami.2021.3057446}, url = {https://doi.org/10.1109%2Ftpami.2021.3057446}, year = 2021, publisher = {Institute of Electrical and Electronics Engineers ({IEEE})}, pages = {1--1}, author = {Matthias Delange and Rahaf Aljundi and Marc Masana and Sarah Parisot and Xu Jia and Ales Leonardis and Greg Slabaugh and Tinne Tuytelaars}, title = {A continual learning survey: Defying forgetting in classification tasks}, journal = {{IEEE} Transactions on Pattern Analysis and Machine Intelligence} }