{"entity":{"id":"self-supervised-pretraining","kind":"term","name":"Self-supervised pretraining (SSL)","aka":["self-supervised pretraining","self-supervised learning","SSL pretraining","label-free pretraining","pretraining objective","pretrained encoder"],"tldr":"Self-supervised pretraining teaches a model from unlabelled data by hiding part of each example and asking it to predict the hidden part, so no expert labels are needed.","summary":"Self-supervised learning trains a model on a task where the data itself provides the supervisory signal rather than external labels (Wikipedia); masking, contrastive objectives and next-token prediction are the common forms. It matters in cancer because unlabelled slides, cells and expression profiles exist by the million while outcome labels exist by the hundred. Whether pretraining beats training from scratch on the labelled task is an empirical question and, for bulk expression at TCGA scale, several groups have reported that it does not.","asOf":"2026-09-24","wikipedia":"https://en.wikipedia.org/wiki/Self-supervised_learning","links":[{"label":"Wikipedia","url":"https://en.wikipedia.org/wiki/Self-supervised_learning"}],"tags":["cansim-terms"],"related":["cancer-ai-vocabulary"],"cancers":[],"sections":[],"technologies":[],"targets":[],"drugs":[],"companies":[],"institutions":[],"pathways":[],"terms":["masked-modelling","contrastive-learning","transfer-learning","foundation-model"],"trials":[],"people":[],"bottlenecks":[],"keyPapers":[],"journals":[],"dependsOn":[],"notes":["Listed in the CanSim terms map 1.0.0 (docs/onco/terms.json, generated 2026-09-24), CC BY 4.0, attribution: CanSim project, an open, public-data-first cancer foundation-model programme; CanSim page path /terms/self-supervised-pretraining."],"provenance":{"editedBy":"OnCo CanSim terms wave (Wikipedia summaries, standards and project pages, GDC and FDA pages, Europe PMC)","editedOn":"2026-09-24","note":"CanSim terms map 1.0.0 (docs/onco/terms.json, generated 2026-09-24), CC BY 4.0, attribution: CanSim project, an open, public-data-first cancer foundation-model programme"},"category":"Methods and models"},"route":"/terms/self-supervised-pretraining/","neighbours":{"term":[{"id":"cancer-ai-vocabulary","kind":"term","name":"Cancer AI vocabulary (CanSim terms map)","route":"/terms/cancer-ai-vocabulary/"},{"id":"contrastive-learning","kind":"term","name":"Contrastive learning (InfoNCE)","route":"/terms/contrastive-learning/"},{"id":"foundation-model","kind":"term","name":"Foundation model","route":"/terms/foundation-model/"},{"id":"masked-modelling","kind":"term","name":"Masked autoencoders and masked gene modelling","route":"/terms/masked-modelling/"},{"id":"transfer-learning","kind":"term","name":"Transfer learning and the low-label regime","route":"/terms/transfer-learning/"}]}}