{"ID":2836891,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.20170","arxiv_id":"2511.20170","title":"AdaCap: An Adaptive Contrastive Approach for Small-Data Neural Networks","abstract":"Neural networks struggle on small tabular datasets, where tree-based models remain dominant. We introduce Adaptive Contrastive Approach (AdaCap), a training scheme that combines a permutation-based contrastive loss with a Tikhonov-based closed-form output mapping. Across 85 real-world regression datasets and multiple architectures, AdaCap yields consistent and statistically significant improvements in the small-sample regime, particularly for residual models. A meta-predictor trained on dataset characteristics (size, skewness, noise) accurately anticipates when AdaCap is beneficial. These results show that AdaCap acts as a targeted regularization mechanism, strengthening neural networks precisely where they are most fragile. All results and code are publicly available at https://github.com/BrunoBelucci/adacap.","short_abstract":"Neural networks struggle on small tabular datasets, where tree-based models remain dominant. We introduce Adaptive Contrastive Approach (AdaCap), a training scheme that combines a permutation-based contrastive loss with a Tikhonov-based closed-form output mapping. Across 85 real-world regression datasets and multiple a...","url_abs":"https://arxiv.org/abs/2511.20170","url_pdf":"https://arxiv.org/pdf/2511.20170v1","authors":"[\"Bruno Belucci\",\"Karim Lounici\",\"Katia Meziani\"]","published":"2025-11-25T10:50:06Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[]","has_code":false,"code_links":[{"ID":606636,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2836891,"paper_url":"https://arxiv.org/abs/2511.20170","paper_title":"AdaCap: An Adaptive Contrastive Approach for Small-Data Neural Networks","repo_url":"https://github.com/BrunoBelucci/adacap","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}