{"url":"/method/single-path-nas","slug":"single-path-nas","name":"Single-path NAS","full_name":"Single-path NAS","full_name_withheld":false,"description_markdown":"**Single-Path NAS** is a convolutional neural network architecture discovered through the Single-Path [neural architecture search](https://paperswithcode.com/method/neural-architecture-search) approach. The NAS utilises a single-path search space. Specifically, compared to previous differentiable  NAS methods, Single-Path NAS uses one single-path over-parameterized  ConvNet to encode all architectural decisions with shared convolutional kernel parameters. The approach is built upon the  observation that different candidate convolutional operations in NAS  can be viewed as subsets of a single superkernel. Without having to  choose among different paths/operations as in multi-path methods, we instead  solve the NAS problem as finding which subset of kernel weights to use in each ConvNet layer. By sharing the convolutional kernel weights,  we encode all candidate NAS operations into a single superkernel.\r\n\r\nThe architecture itself uses the [inverted residual block](https://paperswithcode.com/method/inverted-residual-block) from [MobileNetV2](https://paperswithcode.com/method/mobilenetv2) as its basic building block.","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/1904.02877v1","title":"Single-Path NAS: Designing Hardware-Efficient ConvNets in less than 4 Hours","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/skmhrk1209/Single-Path-NAS-PyTorch/blob/514abca4c0ce350cc920b9b9277790598774829e/models.py#L60","code_snippet_url_on_a_code_host":true,"categories":[{"area":"Computer Vision","area_id":"computer-vision","collection":"Convolutional Neural Networks","url":"/methods/category/convolutional-neural-networks","pwc_aliases":[]}],"n_papers_tagged":5,"archive_num_papers":null,"papers_newest_first":[{"paper":"/paper/balanced-mixture-of-supernets-for-learning","title":"Balanced Mixture of SuperNets for Learning the CNN Pooling Architecture","date":"2023-06-21","arxiv_id":"2306.11982","n_code_links":1,"syntology":null},{"paper":"/paper/s3nas-fast-npu-aware-neural-architecture","title":"S3NAS: Fast NPU-aware Neural Architecture Search Methodology","date":"2020-09-04","arxiv_id":"2009.02009","n_code_links":1,"syntology":{"ran":0,"of":5,"unverified":5,"pointer_only":0}},{"paper":"/paper/single-path-mobile-automl-efficient-convnet","title":"Single-Path Mobile AutoML: Efficient ConvNet Design and NAS Hyperparameter Optimization","date":"2019-07-01","arxiv_id":"1907.00959","n_code_links":1,"syntology":null},{"paper":null,"title":"Single-Path NAS: Device-Aware Efficient ConvNet Design","date":"2019-05-10","arxiv_id":"1905.04159","n_code_links":0,"syntology":null},{"paper":"/paper/single-path-nas-designing-hardware-efficient","title":"Single-Path NAS: Designing Hardware-Efficient ConvNets in less than 4 Hours","date":"2019-04-05","arxiv_id":"1904.02877","n_code_links":9,"syntology":{"ran":1,"of":16,"unverified":15,"pointer_only":0}}],"papers_shown":5,"tasks":[{"task":"/task/image-classification","name":"Image Classification","papers":5},{"task":"/task/architecture-search","name":"Neural Architecture Search","papers":5},{"task":"/task/image-classification","name":"image-classification","papers":4},{"task":null,"name":"GPU","papers":3},{"task":"/task/classification","name":"General Classification","papers":2},{"task":"/task/automl","name":"AutoML","papers":1},{"task":"/task/hyperparameter-optimization","name":"Hyperparameter Optimization","papers":1}],"tasks_shown":7,"n_tasks":7,"usage_by_year":[{"year":"2019","papers":3},{"year":"2020","papers":1},{"year":"2023","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/single-path-nas"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}