We warmly invite all machine learning researchers to participate in the upcoming IJCAI 2021 First International Workshop on Continual Semi-Supervised Learning
https://sites.google.com/view/sscl-workshop-ijcai-2021/ The aim of this workshop is to formalise this new continual semi-supervised learning paradigm, and to introduce it to the machine learning community in order to mobilise effort in this direction. We present the first two benchmark datasets for this problem, derived from significant computer vision scenarios, and propose the first Continual Semi-Supervised Learning Challenges to the research community. To attend, please follow IJCAI guidelines for workshop registration (early bird $80): https://ijcai-21.org/registration-information/ Format ----------------------------------------------------------------------------------------------------------------- The workshop is a full-day event, scheduled in two halves (see below) taking place on Thursday 19th August and Friday 20th August. The event will be hosted on Gather Town, and will also be recorded via Zoom. The programme is a blend of invited talks, oral presentations from the two award-winning papers, two poster sessions and a panel on the future of continual learning. *August 19 – 14:00 – 18:00 UTC* 14:00 Opening remarks 14:10-15:10 Presentation of the benchmarks and challenges 15:10-15:40 Invited talk #1 – Razvan Pascanu 15:40-16:10 Invited talk #2 – Bing Liu 16:10-16:30 *Coffee break* 16:30-17:30 Poster session 1 17:30:18:00 Best student paper award oral presentation *August 20 - 14:00 – 18:00 UTC* 14:00-14:30 Best paper oral presentation 14:30-15:30 Poster session 2 15:30-15:50 *Coffee break* 15:50-16:20 Invited talk #3 – Tinne Tuytelaars 16:20-16:50 Invited talk #4 – Chelsea Finn 16:50-17:50 Panel on future of continual learning 17:50-18:00 Award ceremony and closing remarks Challenges ------------------------------------------------------------------------------------------------------------ With this workshop we intend to propose to the community both a continual activity recognition (CAR) challenge and a continual crowd counting (CCC) challenge. https://sites.google.com/view/sscl-workshop-ijcai-2021/challenges based on a newly released Continual Activity Recognition (CAR) dataset, derived from a fraction of the MEVA (Multiview Extended Video with Activities) activity detection dataset (https://mevadata.org/), and a Continual Crowd Counting (CCC) dataset, derived from sequences from the Mall, UCSD and FDST datasets. Invited speakers ----------------------------------------------------------------------------------------------------- Razvan Pascanu (Deepmind) Tinne Tuytelaars (KU Leuven) Chelsea Finn (Stanford) Bing Liu (University of Illinois at Chicago) Accepted Papers ---------------------------------------------------------------------------------------------------- Lucas Caccia and Joelle Pineau SPeCiaL: Self-Supervised Pretraining for Continual Learning (Best Paper Award) Dhanajit Brahma, Vinay Kumar Verma and Piyush Rai Hypernetworks for Continual Semi-Supervised Learning (Best Student Paper Award) Jiangpeng He and Fengqing Zhu Unsupervised Continual Learning Via Pseudo Labels Luca Monorchio, Marco Capotondi, Mario Corsanici, Wilson Villa, Alessandro De Luca and Francesco Puja Transfer and Continual Supervised Learning for Robotic Grasping through Grasping Features Mahardhika Pratama, Andri Ashfahani and Edwin Lughofer Unsupervised Continual Learning via Self-Adaptive Deep Clustering Approach Enrico Meloni, Alessandro Betti, Lapo Faggi, Simone Marullo, Matteo Tiezzi and Stefano Melacci Evaluating Continual Learning Algorithms by Generating 3D Virtual Environments Qihan Yang, Fan Feng and Rosa H.M. Chan A Benchmark and Empirical Analysis for Replay Methods in Continual Learning Andrea Rosasco, Antonio Carta, Andrea Cossu, Vicenzo Lomonaco and Davide Bacciu Distilled Replay: Overcoming Forgetting through Synthetic Samples Jingbo Sun, Li Yang, Jiaxin Zhang, Frank Liu, Mahantesh Halappanavar, Deliang Fan and Yu Cao Self-supervised Novelty Detection for Continual Learning: A Gradient-based Approach Boosted by Binary Classification Hermann Blum, Francesco Milano, René Zurbrügg, Roland Siegwart, Cesar Cadena and Abel Gawel Self-Improving Semantic Perception for Indoor Localisation Sungmin Cha, Beomyoung Kim, Youngjoon Yoo and Taesup Moon SSUL: Semantic Segmentation with Unknown Label for Exemplar-based Class-Incremental Learning Ajmal Shahbaz, Salman Khan, Mohamad Asiful Hossain, Vincenzo Lomonaco, Kevin Cannons, Zhan Xu and Fabio Cuzzolin International Workshop on Continual Semi-Supervised Learning: Introduction, Benchmarks and Baselines Organising committee ---------------------------------------------------------------------------------------------- Fabio Cuzzolin (Oxford Brookes University) Kevin Cannons (Huawei Technologies Canada) Vincenzo Lomonaco (University of Pisa and ContinualAI) Irina Rish (University of Montreal and MILA) Salman Khan (Oxford Brookes University) Mohamad Asiful Hossain (Huawei Technologies Canada) Ajmal Shahbaz (Oxford Brookes University)
_______________________________________________ uai mailing list uai@engr.orst.edu https://it.engineering.oregonstate.edu/mailman/listinfo/uai
