This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| ICML Workshops | |
|---|---|
| Name | ICML Workshops |
| Discipline | Machine Learning |
| Frequency | Annual |
| Publisher | International Machine Learning Society |
| First | 1980s |
| Country | International |
ICML Workshops ICML Workshops are satellite gatherings held alongside the International Conference on Machine Learning that bring together researchers, practitioners, students, and industry representatives for focused discussion on specialized topics. They foster rapid exchange of ideas, early dissemination of preliminary results, and networking among participants from institutions such as Google, Microsoft Research, OpenAI, DeepMind, and universities like Stanford University, Massachusetts Institute of Technology, University of Cambridge, and University of Toronto. Workshops interface with major events and awards in the field including NeurIPS, COLT, AAAI, IJCAI, and honors like the Turing Award and the NeurIPS Best Paper Award.
Workshops companioning the International Conference on Machine Learning provide focused venues for topics ranging from foundational methods to applied subfields, attracting contributors affiliated with labs such as Facebook AI Research, IBM Research, Amazon Web Services, Apple Machine Learning Research, and institutions including ETH Zurich, Carnegie Mellon University, Princeton University, University of Oxford, and University of California, Berkeley. They operate within the ecosystem of conferences like ICLR and EMNLP, and engage communities represented by societies such as the Association for Computing Machinery and the IEEE. Typical outputs inform subsequent publications at venues like Journal of Machine Learning Research and Transactions on Machine Learning Research as well as datasets and benchmarks from groups like ImageNet, OpenAI Gym, and GLUE.
The workshop tradition at major machine learning conferences traces roots to early specialist meetings connected with events like COLT and AISTATS, evolving alongside institutions such as DARPA and funding agencies like the National Science Foundation and the European Research Council. Over decades, thematic shifts mirrored developments led by figures associated with Geoffrey Hinton-adjacent groups, teams at Google Brain, research from Yoshua Bengio-affiliated labs, and collaborations across centers including Bell Labs, SRI International, Johns Hopkins University, and University of Montreal. The structure adapted from small, invitation-only workshops akin to those hosted by Los Alamos National Laboratory and Bellagio Study Center to open submission formats paralleling evolutions at NeurIPS and ICLR.
Program committees composed of researchers from organizations like DeepMind, Microsoft Research Cambridge, Google Research, Amazon Research, Meta AI Research, and universities such as Columbia University and Yale University set scopes and review standards. Calls for papers and abstracts reference platforms including OpenReview, CMT and repositories like arXiv. Submission types commonly include short papers, extended abstracts, demonstration proposals, and posters with shepherding models influenced by practices at NeurIPS and editorial processes used by journals like Nature Machine Intelligence and Science Robotics. Selecting organizers often involves coordination with the International Machine Learning Society and conference chairs who have affiliations with institutes like UC Berkeley, ETH Zurich, and University of Washington.
Workshops feature formats such as invited talks from researchers at Google DeepMind, panels including participants from Stanford University and Harvard University, lightning talks, poster sessions, tutorials, code sprints, and reproducibility challenges inspired by efforts from Papers with Code and initiatives at ACL and NeurIPS. Activities often include datasets releases by groups like Stanford Vision Lab, shared tasks similar to those organized by Kaggle and runbooks from OpenAI Scholars, and hackathons modeled after events at MIT Media Lab and Berkeley Artificial Intelligence Research.
Prominent recurring themes have included deep learning interpretability sessions with contributors from MIT CSAIL, fairness and ethics panels citing work from ProPublica-adjacent research, reinforcement learning workshops led by teams at DeepMind and OpenAI, and probabilistic modeling tracks featuring researchers from Cambridge University and University College London. Other well-attended areas include graphical models, causal inference influenced by scholars at Harvard University and University of Chicago, optimization and theory echoing contributions from Princeton University and Columbia University, and applications in healthcare with participants from Mayo Clinic, Johns Hopkins Medicine, and Imperial College London.
Workshops enable early-career researchers from programs like CIFAR and fellowships including the Schmidt Science Fellows to present work, connect with industrial hiring pipelines at Google, Meta, Amazon, and Microsoft, and collaborate with labs in the European Laboratory for Learning and Intelligent Systems. They catalyze special issues in journals such as Machine Learning (journal) and serve as incubators for standards and benchmarks that later influence competitions at ImageNet Large Scale Visual Recognition Challenge and leaderboards maintained by Papers with Code.
Common criticisms address proliferation and overlap with symposia at NeurIPS and ICLR, variability in reviewing quality compared with journals and flagship conference tracks like those at NeurIPS and ICML main conference organizers, and concerns about accessibility tied to travel costs borne by attendees from institutions such as University of Nairobi or Tsinghua University. Other debates involve industry influence from companies like Google and Facebook and reproducibility issues discussed alongside efforts from Reproducibility in Machine Learning initiatives and editorial responses in outlets such as Nature and Science.
Category:Machine learning conferences