iCaRL: Incremental classifier and representation learning Rebuffi, Sylvestre Alvise Kolesnikov, Alexander Sperl, Georg Lampert, Christoph ; https://orcid.org/0000-0001-8622-7887 A major open problem on the road to artificial intelligence is the development of incrementally learning systems that learn about more and more concepts over time from a stream of data. In this work, we introduce a new training strategy, iCaRL, that allows learning in such a class-incremental way: only the training data for a small number of classes has to be present at the same time and new classes can be added progressively. iCaRL learns strong classifiers and a data representation simultaneously. This distinguishes it from earlier works that were fundamentally limited to fixed data representations and therefore incompatible with deep learning architectures. We show by experiments on CIFAR-100 and ImageNet ILSVRC 2012 data that iCaRL can learn many classes incrementally over a long period of time where other strategies quickly fail. IEEE 2017 info:eu-repo/semantics/conferenceObject doc-type:conferenceObject text http://purl.org/coar/resource_type/c_5794 https://research-explorer.ista.ac.at/record/998 Rebuffi SA, Kolesnikov A, Sperl G, Lampert C. iCaRL: Incremental classifier and representation learning. In: Vol 2017. IEEE; 2017:5533-5542. doi:<a href="https://doi.org/10.1109/CVPR.2017.587">10.1109/CVPR.2017.587</a> eng info:eu-repo/semantics/altIdentifier/doi/10.1109/CVPR.2017.587 info:eu-repo/semantics/altIdentifier/isbn/978-153860457-1 info:eu-repo/semantics/altIdentifier/wos/000418371405066 info:eu-repo/semantics/altIdentifier/arxiv/1611.07725 info:eu-repo/semantics/openAccess