{"url":"/method/tabularargn","slug":"tabularargn","name":"TabularARGN","full_name":"Tabular Auto-Regressive Generative Network","full_name_withheld":false,"description_markdown":"Unlike synthetic data generators that rely on increasingly complex and resource-heavy architectures, TabularARGN adopts a more **focused and efficient model design**. These design choices result in:\r\n\r\n* **High Fidelity**: TabularARGN achieves synthetic data quality on par with state-of-the-art (SOTA) models\r\n* **Privacy by Design**: TabularARGN only considers privacy-preserving value ranges for sampling, and has built-in privacy protection features. Plus can be trained via DP-SGD for obtaining differential privacy guarantees.\r\n* **Simplicity**: TabularARGN leverages existing building blocks, and thus can be easily implemented within standard deep learning frameworks.\r\n* **Compute Efficiency**: With training speeds up to **100x faster**, TabularARGN scales effectively, even for large and complex datasets.\r\n* **Sampling Flexibility**: TabularARGN supports advanced sampling capabilities, including:\r\n   * **Conditional generation** to create targeted datasets.\r\n   * **Missing value imputation** to handle incomplete data seamlessly.\r\n   * **Fairness adjustments** to align with ethical data synthesis goals.\r\n   * **Controlling sampling probabilities** via temperature adjustments to balance rule-adherence with data diversity.\r\n* **Data Versatility**: TabularARGN accommodates the heterogeneity of real-world tabular datasets, including:\r\n   * Multi-variate, mixed-type data (categorical, numerical, date-time, geo-spatial).\r\n   * Multi-sequence datasets with varying sequence lengths and varying time intervals.\r\n   * Missing values.\r\n* **Robustness in Training**: TabularARGN delivers high-quality synthetic data with default settings and remains consistent across several training runs.","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":null,"title":null,"url_on_a_paper_host":false},"code_snippet_url":null,"code_snippet_url_on_a_code_host":false,"categories":[{"area":"General","area_id":"general","collection":"Tabular Data Generation","url":"/methods/category/tabular-data-generation","pwc_aliases":[]}],"n_papers_tagged":1,"archive_num_papers":1,"papers_newest_first":[{"paper":"/paper/tabularargn-a-flexible-and-efficient-auto-1","title":"TabularARGN: A Flexible and Efficient Auto-Regressive Framework for Generating High-Fidelity Synthetic Data","date":"2025-01-21","arxiv_id":"2501.12012","n_code_links":2,"syntology":{"ran":1,"of":5,"unverified":4,"pointer_only":0}}],"papers_shown":1,"tasks":[{"task":"/task/fairness","name":"Fairness","papers":1},{"task":"/task/imputation","name":"Imputation","papers":1},{"task":"/task/synthetic-data-generation","name":"Synthetic Data Generation","papers":1},{"task":"/task/tabular-data-generation","name":"Tabular Data Generation","papers":1}],"tasks_shown":4,"n_tasks":4,"usage_by_year":[{"year":"2025","papers":1}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/tabularargn"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}