{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/efficient-image-dataset-classification","title":"Efficient Image Dataset Classification Difficulty Estimation for Predicting Deep-Learning Accuracy","arxiv_id":"1803.09588","date":"2018-03-26","proceeding":null,"authors":["Florian Scheidegger","Roxana Istrate","Giovanni Mariani","Luca Benini","Costas Bekas","Cristiano Malossi"],"abstract":"In the deep-learning community new algorithms are published at an incredible\npace. Therefore, solving an image classification problem for new datasets\nbecomes a challenging task, as it requires to re-evaluate published algorithms\nand their different configurations in order to find a close to optimal\nclassifier. To facilitate this process, before biasing our decision towards a\nclass of neural networks or running an expensive search over the network space,\nwe propose to estimate the classification difficulty of the dataset. Our method\ncomputes a single number that characterizes the dataset difficulty 27x faster\nthan training state-of-the-art networks. The proposed method can be used in\ncombination with network topology and hyper-parameter search optimizers to\nefficiently drive the search towards promising neural-network configurations.","url_abs":"http://arxiv.org/abs/1803.09588v1","url_pdf":"http://arxiv.org/pdf/1803.09588v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"efficient-image-dataset-classification","repo_url":"https://github.com/IBM/iotnets","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}