{"url":"/method/singular-value-clipping","slug":"singular-value-clipping","name":"Singular Value Clipping","full_name":"Singular Value Clipping","full_name_withheld":false,"description_markdown":"**Singular Value Clipping (SVC)** is an adversarial training technique used by [TGAN](https://paperswithcode.com/method/tgan) to enforce the 1-Lipschitz constraint of the [WGAN](https://paperswithcode.com/method/wgan) objective. It is a constraint to all linear layers in the discriminator that satisfies the spectral norm of weight parameter $W$ is equal or less than one. This\r\nmeans that the singular values of weight matrix are all one or less. Therefore singular value decomposition (SVD) is performed after a parameter update, replacing all the singular values larger than one with one, and the parameters are reconstructed with them. The same operation is applied to convolutional layers by interpreting a higher order tensor in weight parameter as a matrix $\\hat{W}$.","description_state":"present","introduced_year":null,"introduced_by":{"title":null,"paper":null,"first_author":null,"n_authors":0,"url_abs":null,"archive_paper_url":null},"source":{"url":"http://arxiv.org/abs/1611.06624v3","title":"Temporal Generative Adversarial Nets with Singular Value Clipping","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/pfnet-research/tgan/blob/53fde1bcb59e2a57f76b9b26097226ec0423811f/updaters/tgan_updater_wgan.py#L10","code_snippet_url_on_a_code_host":true,"categories":[{"area":"General","area_id":"general","collection":"Adversarial Training","url":"/methods/category/adversarial-training","pwc_aliases":[]}],"n_papers_tagged":3,"archive_num_papers":null,"papers_newest_first":[{"paper":null,"title":"DTGAN: Differential Private Training for Tabular GANs","date":"2021-07-06","arxiv_id":"2107.02521","n_code_links":0,"syntology":null},{"paper":"/paper/tabular-gans-for-uneven-distribution","title":"Tabular GANs for uneven distribution","date":"2020-10-01","arxiv_id":"2010.00638","n_code_links":1,"syntology":null},{"paper":"/paper/temporal-generative-adversarial-nets-with","title":"Temporal Generative Adversarial Nets with Singular Value Clipping","date":"2016-11-21","arxiv_id":"1611.06624","n_code_links":4,"syntology":{"ran":0,"of":3,"unverified":3,"pointer_only":0}}],"papers_shown":3,"tasks":[{"task":"/task/attribute","name":"Attribute","papers":1},{"task":"/task/image-generation","name":"Image Generation","papers":1},{"task":"/task/tabular-data-generation","name":"Tabular Data Generation","papers":1},{"task":"/task/video-generation","name":"Video Generation","papers":1}],"tasks_shown":4,"n_tasks":4,"usage_by_year":[{"year":"2016","papers":1},{"year":"2020","papers":1},{"year":"2021","papers":1}],"row_source":"embedded","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/singular-value-clipping"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}