{"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/reachability-analysis-of-deep-neural-networks","title":"Reachability Analysis of Deep Neural Networks with Provable Guarantees","arxiv_id":"1805.02242","date":"2018-05-06","proceeding":null,"authors":["Wenjie Ruan","Xiaowei Huang","Marta Kwiatkowska"],"abstract":"Verifying correctness of deep neural networks (DNNs) is challenging. We study\na generic reachability problem for feed-forward DNNs which, for a given set of\ninputs to the network and a Lipschitz-continuous function over its outputs,\ncomputes the lower and upper bound on the function values. Because the network\nand the function are Lipschitz continuous, all values in the interval between\nthe lower and upper bound are reachable. We show how to obtain the safety\nverification problem, the output range analysis problem and a robustness\nmeasure by instantiating the reachability problem. We present a novel algorithm\nbased on adaptive nested optimisation to solve the reachability problem. The\ntechnique has been implemented and evaluated on a range of DNNs, demonstrating\nits efficiency, scalability and ability to handle a broader class of networks\nthan state-of-the-art verification approaches.","url_abs":"http://arxiv.org/abs/1805.02242v1","url_pdf":"http://arxiv.org/pdf/1805.02242v1.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":"reachability-analysis-of-deep-neural-networks","repo_url":"https://github.com/trustAI/DeepGO","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"paper_slug":"reachability-analysis-of-deep-neural-networks","repo_url":"https://github.com/Accountable-Machine-Intelligence-Lab/DeepGO","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.02242","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}