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.
| ALT (workshop) | |
|---|---|
| Name | ALT (workshop) |
| Genre | Workshop |
| Location | Various |
| First | 2000s |
| Organizer | Academic and industry consortia |
| Participants | Researchers, practitioners, students |
ALT (workshop) is an interdisciplinary workshop series focused on algorithmic learning theory, adaptive learning technologies, and applied learning techniques that brings together researchers from computer science, statistics, cognitive science, and engineering. The workshop convenes scholars and practitioners to present new methods, compare empirical results, and foster collaborations among universities, research labs, and technology companies. Participants often include faculty from leading institutions, researchers from national laboratories, and engineers from major technology firms.
The workshop foregrounds topics such as computational learning theory, statistical learning, online learning, reinforcement learning, and representation learning, attracting contributors connected to MIT, Stanford University, University of California, Berkeley, Carnegie Mellon University, University of Cambridge, Oxford University, ETH Zurich, University of Toronto, Princeton University, Harvard University, Yale University, Columbia University, University of Washington, University of Illinois Urbana–Champaign, University of Michigan, University of Pennsylvania, Caltech, Imperial College London, Tsinghua University, Peking University, University of Edinburgh, University of Tokyo, Seoul National University, National University of Singapore, Australian National University, University of Melbourne, McGill University, University of British Columbia, École Polytechnique Fédérale de Lausanne, Max Planck Society, Google, Facebook, Microsoft Research, DeepMind, OpenAI, IBM Research, Amazon Web Services, NVIDIA, Intel, Huawei, Baidu Research, Apple Inc., Samsung Electronics, Tencent, Alibaba Group.
Contributions at the workshop often relate to foundational results like PAC learning, VC dimension, boosting, bandits, and kernel methods, with ties to influential work from researchers affiliated with Leslie Valiant, Vladimir Vapnik, Andrew Ng, Geoffrey Hinton, Yoshua Bengio, Michael Jordan, Shai Shalev-Shwartz, Stefano Soatto, Peter Dayan, David Silver, Ian Goodfellow, Fei-Fei Li, Daphne Koller, Judea Pearl, Tom Mitchell.
The workshop originated in the early 2000s as a response to growing interest in algorithmic and adaptive approaches across machine learning subfields, building on earlier venues such as the COLT conference, the NeurIPS workshops, and meetings associated with the Association for Computing Machinery and the Institute of Electrical and Electronics Engineers. Early editions showcased collaborations among groups from Bell Labs, SRI International, Microsoft Research, and national research centers like the Lawrence Berkeley National Laboratory and Los Alamos National Laboratory.
Over time the program expanded to incorporate topics from deep learning, probabilistic modeling, causal inference, and human-in-the-loop systems, drawing speakers from institutions behind landmark projects such as ImageNet, AlphaGo, BERT, GPT, ResNet, Transformer (machine learning model), and initiatives at DARPA, European Research Council, and National Science Foundation. The evolution reflects influence from major workshops at ICML, AAAI Conference on Artificial Intelligence, KDD, AISTATS, UAI, EMNLP, and CVPR.
Typical formats include keynote lectures, paper presentations, poster sessions, panel discussions, and hands-on tutorials. Keynotes often feature senior researchers from Google DeepMind, OpenAI, Microsoft Research, Facebook AI Research, IBM Watson, Toyota Research Institute, Siemens, Siemens Healthineers, or academic chairs from Stanford AI Lab, Berkeley AI Research, Oxford Machine Learning Research Group. Tutorials cover methods from stochastic optimization, convex analysis, kernel methods, to variational inference, Monte Carlo methods, causal discovery, and multi-armed bandits, referencing seminal approaches developed by figures associated with Yann LeCun, Trevor Hastie, Robert Tibshirani, Bradley Efron, Leo Breiman, Jerome Friedman.
Interactive activities include code sprints, reproducibility challenges, benchmark evaluations derived from datasets like MNIST, CIFAR-10, ImageNet, COCO, GLUE, SQuAD, and competitions modeled after platforms like Kaggle and collaborative reproducibility efforts akin to Open Science Framework.
Organization typically involves program committees composed of researchers from universities and industry labs, with organizing institutions including departmental units and centers such as Center for Data Science, Machine Learning Group at Cambridge, and consortia supported by funding agencies like the National Science Foundation, European Commission, UK Research and Innovation, Japan Society for the Promotion of Science, and private sponsors among Google.org, Microsoft Philanthropies, Intel Labs, Amazon Science, NVIDIA Research, Facebook AI. Local hosts have included academic departments at ETH Zurich, University of Toronto, University of Oxford, Imperial College London, Tsinghua University, Peking University, University of Melbourne, and research parks such as Silicon Valley incubators and innovation hubs in Boston and Seattle.
Program chairs often coordinate with editorial venues and proceedings publishers like Springer, ACM, IEEE, and preprint distribution through arXiv.
The workshop has been cited for accelerating cross-pollination among theoretical and applied strands of learning research, influencing follow-up work presented at ICLR, NeurIPS, ICML, AAAI, and KDD. Outcomes include novel algorithms, improved evaluation protocols, and community-driven benchmarks that informed products and deployments at companies like Google, Microsoft, Amazon, Apple, and research breakthroughs referenced in awardees of prizes such as the Turing Award, NeurIPS Test of Time Award, and recognitions from national academies like the National Academy of Sciences and Royal Society.
Critics and commentators in outlets tied to Nature (journal), Science (journal), Communications of the ACM, and professional blogs have debated topics raised at workshop sessions including reproducibility, ethics of deployment, and societal impacts of adaptive systems, with panels sometimes including representatives from ACM FAccT, Partnership on AI, AI Now Institute, and regulatory bodies such as the European Commission's digital policy units.
Selected notable editions have featured presentations and follow-ups that contributed to developments in online convex optimization, bandit algorithms, meta-learning, few-shot learning, causal representation learning, and scalable variational methods. Specific outcomes include influential tutorials later expanded into textbooks and monographs by authors associated with MIT Press, O'Reilly Media, and course materials adopted at Coursera, edX, and university curricula at Stanford University, MIT, Berkeley. Collaborative projects seeded at workshops have resulted in open-source releases under organizations like the Apache Software Foundation, TensorFlow, PyTorch, and benchmark suites incorporated into industrial workflows at Google Research and Facebook AI Research.
Category:Workshops